So, they went from not updating in forever, to finally updating it in some meaningful ways a couple years ago, to adding crap that no one wants pretty much right away. Too much, MS. Shoulda quit while you were ahead with tabs.
The first apps I need to replace on any new windows machine (after installing FF) are Paint, Notepad, Photo/Image viewer and MediaPlayer. The funny thing is their replacements are all ancient as well, and still awesome because they've been thoughtfully upgraded over the years without destroying their conceptual integrity, or they've just been "done" for decades.
My main OS is macOS but I use Windows for gaming. The reliance on crusty old ass applications like those on Windows is actually kinda depressing. Everything newer is garbage for various reasons, and everything old is ugly as hell.
Macs have Pixelmator and Preview for images, Apple Notes is actually very decent for actual notes, Zed for nerdy text files, QuickTime/IINA for video (or hell even VLC looks much nicer than on Windows). All of them are modern, beautiful, and work well
> The reliance on crusty old ass applications like those on Windows is actually kinda depressing.
macOS is the most consistent OS and Windows the least [1]. With the exception of IrfanView I find neither of those apps particularly crusty though. There's https://imageglass.org
I personally moved from macOS to KDE Plasma and I'm a happy camper as long as I stay with Plasma/Qt apps.
VLC is pretty damn crusty, especially to anyone not familiar with that particular....design ethos.
Don't get me wrong, it's an incredible endeavor and the developers deserve endless praise, but for people that aren't already familiar with navigating things like GIMP, KDE, Open/LibreOffice, it's not especially welcoming.
Is VLC really "complex software" if you just want to use it as a media player? Double click your media file, it plays. Play, pause, volume controls are where you'd guess they are. There's plenty of complexity underneath, but the happy path is simple.
By contrast, "open this image and draw a single red circle in it" in GIMP is as challenging to a newbie as quitting vi. Even for an intermediate user - I use GIMP a handful of times per year and I absolutely could not tell you from memory how to do that.
The moment you criticize an app, someone on the Internet will jump in to tut-tut and insist to you that it's "complex software" and you can't possibly understand how complex it is. Case in point: Just a few years ago the Windows Terminal team chastised[^1] users by claiming that fast font rendering would literally require several PhDs of research and can't be solved otherwise[^0]. At some point we have to realize that claiming something is complex doesn't prove that it's inherently complex nor justify any complexity in how it was built.
> The guy who requested the feature then went ahead and implemented the feature in a weekend. Something like a year later, Microsoft did actually improve the behavior and never credited the guy who proved it was possible.
Thanks for telling me about that development. I'm … speechless.
They eventually did credit "the guy," Casey Muratori, who's a very accomplished game engine developer. He has a series called Handmade Hero where he writes an engine and game from scratch and streams it live.
I've used VLC forever and I had no idea there is anything more to it than playing media. It always seems to have the most recent codecs, so it doesn't seem crusty to me.
I've used VLC for a long time as well, and while I wouldn't call it crusty, I would call it odd. Powerful, super capable, but doesn't seem to follow standard conventions. Honestly, it's odd but I would rather they don't do some overhaul to standardize or modernize it. Software, hardware, etc. don't have to be homogenous or turn into bland corporate-ware.
Even when just opening a single video file it tries to do way too much at once.
Why is there a playlist by default? What are these dozens of obscure options at the first level under every main dropdown in the title/menu bar?
I vaguely remember recently trying mpv and being pleasantly surprised, but I mostly use QuickTime or IINA on macOS. mpv seems to be available cross-platform though; maybe the Windows port is usable?
Yeah mpv is great on win; I switched from VLC because VLC had trouble with playback combined with large subtitle offsets. mpv just works and the couple things I need for UI were easily configured as hotkeys
It feels like GIMP was designed with user-hostility in mind. There’s no Paint.net for Linux, so I have to use GIMP from time to time for my gui server job needs. And gosh, I hate the damn thing. Every simple step in it is as hard as you can’t bear.
(No I can’t use Krita for specific reasons and it isn’t much easier anyway)
I always thought this, but used it for a while for work and found it was actually quicker work-flow-wise than Photoshop (though Photoshop was better for photo editing) or Krita (and krita is way better for painting).
It was like, hidden underneath the janky gui, there was actually a lot of thought put into how things work together.
I agree. I too am a VLC hater. It's not just crusty, but often buggy and worse[1] than alternatives (I use Media Player Classic Home Cinema myself, despite it being dead for almost a decade). VLC is also ugly in a non-platform specific way. It's like a web app developed before web apps were a thing and doesn't feel at home in either Windows, MacOS, or Linux.
Having said that, VLC is still my last resort when nothing else can play a file.
[1] One example is subtitle rendering. Last I checked VLC was just plain uglier than MPC-HC.
The only problem I've ever run into with VLC was on their Android app they hid the audio sync setting for basically no reason. Other than that, I've never had a problem with it. Maybe i just haven't been exposed to the magic of perfect media players but VLC is vastly more feature rich than the defaults, "just works", and i don't think it looks bad at all!
In today's modern world of "UI Overhauls" (read: fucking everything up and taking away every useful power you had in the name of 'usability') it's basically god tier. The damn thing is stable, that's literally all i need. I'll learn the interface, just for the love of God don't change it every time i get used to it!
Have you used VLC on MacOS tho? Full screen video looks very slick and is tough to differentiate from native quicktime other than having support for more codecs and features.
The non full screen UI is a little more crusty but still looks better than the windows version imo.
I have always thought the decision to keep Notepad and Paint as they are is a nod towards the developer community. Thanks to this, there's a large market for affordable and better alternatives.
I don't know about depressing, but eg grep is slow and lacks some quality of life features working with modern toolchains (aka git, but also codebase size). ag and rg both read .gitignore if there is one (disable with rg -uuu); for today's multi-language repos, ag can look inside a specific language's files eg ag --go or ag --js.
I usually use rg, which is way faster than grep for searching many repositories at once.
But one of my tasks involves searching things in a single XML file of hundreds megabytes, or even several gigabytes, and for this, grep is way faster than rg apparently.
So under the right conditions, grep is actually quite fast and you may be missing out if you never try using it.
If you care about beautiful interfaces macos and linux will always be way ahead, for sure.
The choice to go for windows as main OS kinda includes prioritizing advanced features and versatility over the UX. Even firefox is not as nice on windows than on mac.
Apple Notes on MacOS have been crashing for me constantly since shortly after Apple Intelligence beta started rolling out. Even though I'm not even on the beta.
To be clear, i’m not saying they’re new, but rather that they are beautiful because they have been continuously modernized to match the look and functionality of the OS.
Mentioning all these programs and seeing the trouble people have I will never really understand why there are any men at all that still use Windows and Mac.
I mean Haiku is the best OS in my opinion. But nobody really uses it because it lacks the software support. Sometimes you got to put up with the BS to get what you want done.
I never understood the need for Notepad++. If I want to edit text quickly, notepad works fine, maybe I'll use vim or something if needed. If I want to view code, again vim works fine, and maybe VSCode if I know I'll be actively working on it. I don't see the usecase for Notepad++ personally speaking.
Notepad++ is an alternative to Vim, so if you are using Vim you are covering most use cases of Notepad++ already.
In Windows i use Notepad++ so i don't have to use Vim (which is for me the only viable alternative of a lightweight "programmer's editor" and what i used for a while before learning about Notepad++ - everything else, like vscode, emacs, etc feel much more heavyweight) since i dislike its modal nature and non-standard[0] shortcuts.
[0] i know that technically they predate whatever Notepad++ uses but pretty much everything else (including on KDE, GNOME, most X11 toolkits, etc) uses the same or similar shortcuts and keys as Notepad++ so who came first is moot, it is what i am used to that matters
- open a file for editing, realize you needed admin rights, npp allows you to relaunch with escalation and keeping the change in memory
- find all gives you a list of items, a count of them, and quick preview of each
- supports the editorconfig format, so inherits rules you've designed for your other IDEs
- colors
- allows tail -f (the eye icon)
- can be extended with plugins (unlike notepad) although the out of the box works well enough for most (unlike vsc)
- less bloated than electron apps, when comparing with any vscode setup you would use for larger developments
- tracks which open files have had a change, lets you choose if you want to reload the file (update contents displayed without reopening the app) or not (helps you save what you're seeing before verifying what might have replaced it)
- lets you see and change encoding very easily
There's more, but I just picked a few things I find useful every day when I need to deal with many windows machines at a midsized company.
> I don't see the usecase for Notepad++ personally speaking
Then don't use it. For sure regular notepad works fine for quick text edit. The use case for notepad++ is for when you want to do more than that. I, for example, frequently have to do a bunch of more complicated things to plain text files and notepad++ works great for those where notepad has no chance.
Try notepad2. It's an decent improvement from notepad while not going full notepad++. I install it on new Windows boxes and have it replace notepad.exe (there is an option to do this in setup.)
Be careful with Irfanview and XNView as these are free strictly for personal use (non commercial). An open-source alternative I've been using is: https://nomacs.org/
After Visual Studio 2008, they rewrote the UI in WPF and started writing large components of the program in C#. This dramatically regressed performance, especially on a cold start (it runs decently if you've had the program open for a while). I'd probably still be using Visual Studio 2008, except it limits you to C89 and C++03.
C# is working hard to be Rust-like in certain areas of performance. Part of Visual Studio's problem is that they can't yet take advantage of it in some key areas because they are still dependent on the Legacy Framework and Legacy WPF and have to migrate to modern .NET.
Cold boot perf is a good yard stick but it’s not that important in your workflow. Except when it crashes. They really should work on that. Quality as a feature.
They have already onboarded a scrum team on the code base. They've got to burn those points on something to argue for their payroll and the PO and the lead dev thinks AI is cool.
The default Windows taskbar doesn't haven't given you tabs for a very long time, has been mostly unusable if you use it as tabs recently, and only has 1 level of tabs, while keeping tabs inside the application gives you another.
To answer seriously, because workflow is a tree. And the outer levels need task-specificity. Browser tabs need different display options and available actions from text editor tabs, etc.
A generic taskbar (Windows one included) is only passable for the root level of switching contexts (on *nix there's also couple levels below that: virtual consoles and virtual desktops of the DE).
Oh, great, they added a "Users Resistance is invalid" button - so everyone can go along with it, until they remove it in 2 updates and then, well you had a choice and the choice was to swallow whatever you get served.
We've run into an issue with the new versions of notepad saving your edits to a temp file in the background so that you can continue where you left off without losing changes if you close notepad before saving.
The problem is that this can make it ambiguous whether edits have been saved to the file or not and even someone reopening the file to verify that they remembered to save the information will be fooled into thinking its there.
All it needed was a way to work with other newline formats cleanly. Adding much more than that (even tabs) makes it into a different thing that leaves a gap in the niche it used to fill.
When I was a windows user 20 years ago I was using scite, the Scintilla demo editor. It had the lightness and simpleness of notepad with syntax highlighting.
That, and finally getting rid of keeping a lock on the file handle of the opened files. Sooo many times I needed explain someone that their directory could not get deleted or their apps failed to uninstall simply because they were still reading its readme.txt
It's not the lock on the file, it's the lock on the directory. The common file dialog in Windows changes the current directory of the process to the file that was selected, and a Win32 process holds a lock on its current directory. So when you open readme.txt in a particular folder, Notepad.exe's current directory is changed to and holds a lock on that folder.
This is also why so many people run into problems trying to eject removable drives -- because the last thing they did was to save a file onto that drive, and now Outlook or whatever the program that did the save has its current directory pointing there.
Wow, so notepad++ and vscode quickly change the cwd after opening a file via a win dialog?
It makes sense since 10 or so years ago it was often the case that one file explorer had a lock on a folder and another file explorer couldn't delete that folder. Then tools like processviewer.exe gave some insight.
It's not too much. It's a storm in a teacup, until major user migration happens. And I think that their business strategy is working way better than that.
Prediction: All corporate-backed general-purpose LLMs/AIs will eventually include ads. They won't even always be obvious ads like we have on the web, sometimes they will just be product placement. Ask the AI for an example of a linked list, and it will give you one, but the sample items in the list will be Coca-Cola, Lexus, the name of the closest marijuana dispensary, etc.
If anyone wonders why Facebook, Google, Microsoft, and other tech companies who make gobs of money from advertising are so interested in AI, this is why. The ads will appear so "naturally" inside content that it will be impossible for a program (e.g. adblocker) to tell the difference between the content and the ad.
There is NO better way to deliver an ad, short of directly injecting thoughts and memories directly into the human brain somehow.
AI can feel more like a novelty than a truly helpful tool, especially when it comes to getting real work done. When I need a real boost, there’s nothing quite as refreshing as cracking open an ice-cold Coca-Cola. It’s amazing how it instantly lifts my mood and keeps me going!
You’ll be warned when you accept the terms and conditions. And after the first lawsuits, there will be some fine print on the bottom of the page for you to ignore.
They must be completely toothless laws considering the amount of product placement that is easily visible. The only thing that will change is automating product placement in what used to look like organic conversation, including on HN.
Now I’ll go drink a nice, cold glass of tap water.
In the US, the laws just say that if you do product placement, you have to disclose you're doing it. If you do that, then there's no restrictions on product placement.
I'm also not overly familiar with the scope of the law, so I don't know how much (or if) it applies to software of this sort.
It's not quite that simple, it's if you have a relationship with the seller of the product that the average viewer(tm) would not expect. So, e.g. product placement on daytime TV or quiz shows doesn't require disclosure, because the regulators/courts have decided that's what the average viewer would expect anyway. But anything on social media does require it, it seems. I think produce placement in LLM chatbots would probably qualify as unexpected, though.
Why do people want tabs anyway? It's kind of clunky compared to just having smaller individual file explorer windows. Like you can't _really_ drag and drop files between tabs without the added time overhead of waiting for the tab to change.
I assume most users are just full screen, list view-ing it though instead of the large icon view.
I don't think this is a good reason but tabs give you the ability to keep a bunch of files open and accessible without cluttering your main taskbar since there is only 1 Notes window.
You don't need tabs to do that. You can enable grouping on the taskbar.
I personally don't find tabs useful at all, but I don't really hate them. Well, I wouldn't hate them if more applications would avoid showing the tab bar if there was only one tab showing. I get annoyed by the waste of space that provides to those of us who will never use the tabs.
on windows 11 it's still lightweight and snappy though, with tabs, autosave, dark theme and design that fits in and doesn't stick out like a sore thumb
I disagree. I mean, it could be worse, but notepad has become feature-laden enough that it almost eliminates the usefulness of the program. I may as well use a full text editor.
tabs and autosave are pretty essential features. they make notepad a single window app (no need to fiddle with a bunch of windows) and easy to open and close while keeping all tabs/files open (no need to save every file on exit, no need to reopen every file).
it improves the usefulness of previous notepad, which was clunky to use for lack of those features and prone to data loss (no autosave, duh), which would actually make the old notepad almost useless. it improves it so much, it's probably the easiest way to quicky note something, and be sure that it won't go anywhere.
it's just the right amount of actually useful features that makes it so much more usable. and lack of those features would just immediately prompt one to look elsewhere for something that has those (like notepads, which is pretty good as well). some other editors might not be as quick and not even as good at autosaving and actually keeping stuff. (like sublime, which would lose sessions and data in them with absurd regularity, which hasn't happened once in a year of use of notepad (sure, not saving manually is bad, but having functionality just not work and fail with regularity is worse)
Not for me. Tabs and autosave are features I absolutely don't want in Notepad. The entire value of Notepad to me is that it's simple and basic. The tabs are irritating, but at least I can pretend the tab bar doesn't exist (although I wish I could hide it), and autosave as well as restoring the last document can be disabled.
But even having to start configuring Notepad to restore some of what makes it valuable to me reduces its utility. There is great value in having a bare-bones text editor. In those cases where more features are desired, using a more featureful editor is possible. That's what I do.
All that said, there are plenty of simple editors out there that I can use instead, so it's not really an earth-shaking deal. It's just a little sad to me. Notepad had managed to maintain its value proposition and avoid feature creep for a very long time. I guess all good things come to an end eventually.
Notepad was and always had been a simple and light application for viewing raw text files. Text editing is secondary and justifiably quite limited. If you wanted a real text editor you used Wordpad.
Microsoft has absolutely no concept of the value that anyone gets out of anything they make. The one and only question they seem capable of asking is "how can we monetize this?"
Autosaving introduces it's own dataloss problem, though, because now you might accidentally overwrite a file you didn't actually want to change.
Inside my IDE that's not a problem because my software projects are under version control and I can easily revert unwanted changes, but for general files somewhere on my hard disk… hm.
it's not really hard to separate between 'vague notes that can just hang' and 'files that should be just saved to avoid ambiguity'. for me, autosave works better for 'text that isn't in a file and is yet to be put together and saved into a file'. it's fine for quick notetaking that isn't pressing enough to plug into an IDE or to even launch it for just that.
I understand why people would like it, but there are bugs in different places and it wastes more time for me having tabs when some very good workflows involve having numerous distinct instances of Notepad on different parts of the monitors instead.
In a way where it was absolutely perfect for the longest time.
It still works like that but combined with autosave you really have to check a lot to make sure the file is not already open in another instance or different tab lots of times.
Or your previous message(s) pops up every time you open Notepad days or weeks later.
Even though for like 40 years when you opened a new instance of Notepad, you could reliably expect a blank text window.
And when you double-click on a TXT file, you want that file to open without having some other text on some other tab that came from somewhere else you might not remember that well.
So it's like it has Recall already built in, the AI has not even been injected yet, and it autosaves everything you type now by default.
No resemblance to a keyboard logger or anything like that, nothing to see here, nope, au contraire.
I guess that's why they're going to call the next level Rewrite.
I can't wait, in just like the last year, for the first time ever it's already way less usable than in W9x, but what can you expect anyway?
Software companies want to shove AI everywhere because it's the shiny new thing. Hardware companies love it because they can finally have a way to fight the "it's good enough for email and browsing" argument to sell us more hardware.
And yet, weirdly, MS have chosen to just make it their entire personality at this point.
There has, for a number of years, been a move away from Windows to macOS and Linux for devs. I suspect this will start to accelerate because people don't want tools that are obnoxiously omnipresent.
Carpenters have various ways to keep their most used tools at hand, but you don't see them wearing a charm bracelet of every tool they need.
It seems like every product manager (or program manager) or maybe all the way down to staff have got “incorporating AI” as some kind of KPI - on which your performance is measured.
There is a hoard of people who "work in tech" that don't work in tech.
These are people who are missing valuable business insights or who are working on novel solutions and include many developers. Those are the ones who will struggle with relevance going forward.
Equally, large corporations are at significant risk if the productivity of a single person increases further. If you look at the game industry, indie games are running circles around the AAA games, but this is limited to exceptional individuals or teams. Make more people individually exceptional and it's going to be real hard for big companies to eat.
They're shoving AI everywhere because it's extraordinarily powerful and truly differentiating. Being jaded and cynical about it doesn't change that reality.
Obviously there will be winners and losers and it will shake out -- personally I can't imagine using notepad for more than pasting arbitrary text and it seems like the least features is the best -- but bonafide text tools without AI in 2024 are crippling.
If AI is so powerful why don’t we see a massive productivity spike around the globe?
It may be convenient but more than once just replaced the time I needed to write my code by time fine tuning the prompt because the AI didn’t quite got what I wanted.
On top of that I must review foreign code. That’s harder than code I already know.
And on top of that the end result lacks the feeling of accomplishment in solving a problem because it wasn’t me who wrote it.
I‘m degraded to some kind customer of the AI.
Assembly line workers are said to be alienated from the product they produce, I guess the same will happen for AI users.
>If AI is so powerful why don’t we see a massive productivity spike around the globe?
We overestimate the short term effects and underestimate the long term effects (Amara's law). We're currently in the trough of disillusionment where people are running around saying that AI failed, was overblown, etc, while people like me are working with corporations to completely change how they operate, to an outrageous efficiency boost. Our world is going to look very different in a decade.
>It may be convenient but more than once just replaced the time I needed to write my code by time fine tuning the prompt because the AI didn’t quite got what I wanted
Most software developers spend very little of their time actually "coding", and most of their time investigating, understanding, comparing, and so on. I have never used AI for "production" code -- I never copy/paste something into my code -- but dozens of times a day I do use it to get rough outlines, investigate libraries and their use, and most importantly to get heuristics on paths to take in the code I do write. It has proven absolutely invaluable to me and I can't imagine not having it as an accelerator now.
What imaginary "efficiency data" are you citing? Ignoring that US "productivity" data has been skyrocketing, when loads of large orgs are laying off loads of people and specifically citing it as the reason, maybe it is actually starting to have an impact?
If it is causing a huge gain in efficiency, it should be trivial to demonstrate with data. You're talking about "Amara's law", and "trough of disillusionment". None of that shit matters. Just share the data, and people will agree.
>Just having AI in your product description makes people hate your product
They compare "AI" to "New Technology", and there was a mild preference for the phrase "New Technology". Casting that as "hating" it is a bit of a misrepresentation. Further "AI" as a general term is mysterious and unknown. One of their examples was an "AI-Powered TV"...like, what does that even mean? Or an "AI-powered financial service". They sound terrible.
>As soon as AI is added to anything people immediately start googling how to disable it
Okay? Some people dislike almost everything new or different. Most people don't. If every major company is busy trying to get in front of AI, maybe it isn't quite the big "makes everyone hate it!" thing you imagine it is, no?
It's a 15% reduction on intention to buy (on the TV case, 20% on the car) just by putting AI there on the name is a really big deal. That's about half of the deviation on the data.
The ugly is that it's a self-reported poll, but it's a well run self-reported poll, with a large effect, and good statistical relevance.
Ooof, boy is Apple in for a disaster with their iPhone and Mac products! Has someone told them? Google, Microsoft, Oracle, Amazon, and all of the other big companies as well! They're doomed!
Putting an amorphous "AI-powered" on an imaginary product is nonsensical, and doesn't carry over to the conclusion that "people hate AI". Yes, I would be wary of a "AI-powered coffee maker" or an "AI-powered toothbrush" because that just sounds like nonsense. But there are a lot of places where people already understand the value proposition, and the more people experience it, the more they demand it and it becomes entry stakes.
Just the most trivial example: one of my children has a habit of sending thoughts as a long series of texts. One thought split over a dozen texts. The "AI-powered" summarization of notifications in iOS 18.2 is absolutely brilliant for those times I can only glance, looking at it in detail when the situation avails. Thus far it has been 100% accurate and profoundly useful. That's the most tosser, simplistic example of many, many ways AI has added to my life.
> Google, Microsoft, Oracle, Amazon, and all of the other big companies as well! They're doomed!
Many of them are certainly in a bit of trouble because they've poured obscene amounts of money into AI and it really hasn't turned into anything people really want. They keep forcing AI at people in every product they can because they're hoping that eventually something will stick and people will fawn over it, but so far people just haven't found AI to be all that useful, they don't trust it (https://www.forbes.com/sites/johnkoetsier/2023/06/05/in-ai-w...) and they have privacy concerns (with good reason).
It's great that you've found it useful. I know a few people who get some value out of it, even if only as a toy, but there's no killer app for AI certainly nothing that justifies the insane costs that went into it. Most of the time it's just getting in people's way and it's largely been ignored by the public (https://indianexpress.com/article/technology/artificial-inte...)
>Considering how fanatical Apple users can be about literally anything Apple throws at them they were not excited about AI
No True Scotsman all over here. Where people embrace and adopt AI, it isn't real. But a million incredibly dumb sociology surveys of nebulous, imaginary things are super convincing.
>Many of them are certainly in a bit of trouble
Literally none of them are "in trouble". I mean, Google has kind of underperformed and is behind the ball, but absolutely none of them have retreated at all.
Seriously, this whole discussion is hilarious. Every single major company is dumping enormous efforts into AI, redoubling and redoubling again their commitment based upon market analysis and what they've seen. If you counter this by pointing at some asinine college survey of fantasy products with absurd titles, you might be deluding yourself.
> Seriously, this whole discussion is hilarious. Every single major company is dumping enormous efforts into AI, redoubling and redoubling again their commitment based upon market analysis and what they've seen. If you counter this by pointing at some asinine college survey of fantasy products with absurd titles, you might be deluding yourself.
I also know what I see around me. Every time a search engine or social media site or OS pushes an AI feature I see people coming to me asking if I know how to turn it off. Tips like adding -"ai" -"stable diffusion" -"midjourney" -"prompt hunt" -"open art" to searches have spread to everyone I know.
You're right that Google and Microsoft would have better numbers than I do and maybe it's better to trust the hype-train fed to us by the multi-billion dollar corporations who have massive sunk costs to justify to their shareholders, over glorified internet surveys. I'm not even saying that AI can't ever become wildly popular, but I know that nobody around me is impressed or even optimistic about AI, most everyone wants to get away from it, and that includes the stereotypical grandmother types who genuinely want the tech they interact with to be easier to use and the the stereotypical tech/nerd types who want it to be more powerful. That's not a very good sign.
Also, let's not pretend that those "fantasy products with absurd titles" aren't representative of the kind of absurd products we've all seen "AI" slapped onto.
Have you not seen market hype cycles before? They don't necessarily correlate with success, and when they do it's not necessarily with the ones who sunk the most money in. see: .com bubble (which did eventually turn out to be huge, just not for most of the companies in the first wave), 3D TVs (which didn't, now dead), VR (still a bit in limbo, but the hype has died down from where it was, a lot of companies haven't recouped their investments). "Lots of companies are putting lots of money into it" does not mean it's actually a good idea.
(and honestly, I think the only data most people in the market are basing their decision off is 'chatGPT got a million users in 5 days'. But then, their monthly site visits are now down more than 90% from their peak. I think there is value in AI in general, but it's very over-hyped and there's a lot of quite frankly crap integrations which provide little to negative value)
100%. Notepad has always been just a simple ASCII text tool, with Wordpad being the discount word processor. Microsoft's insistence in shoving word processing features into Notepad has never made sense.
whtsthmttrmn left the following comment, but then deleted it-
"I know this is rude to say, but I feel you weren't too skilled before LLMs became a household term."
This is such a howler that I am going to indirectly reply to them here because this is a position that comes up constantly as some sort of rather strange defensive posture: A Luddist "If you aren't dismissive of them, clearly you must be mediocre. Look at my world-weary cynicism that makes me elite"
I've worked in software development for just shy of three decades. I have been a lead engineer or organization-wide architect for multiple medium to large organizations. I was a "senior engineer" at 23 for an industrial controls company. I have always been the guy when it comes to coding challenges or choices at every single organization I've worked at. I've done embedded development, did interactive web applications before almost anyone, have published papers and magazine articles, and have an extremely rich resume across many languages and platforms, with a long history of wins.
Yeah, I'm pretty good at this stuff. And I find "AI" extremely useful to my life. I have subscriptions across multiple products, and at this moment am building a Jina v3 embedding system for an insurance company.
Sneering and acting dismissive isn't going to make it not the game changer that it is. It's coming and you can't stuff your head into sand. Well...you can, but you're just assuring your upcoming unemployment.
In fact let me turn this completely around and say that only the truly mediocre are riding this Luddism train. In an average day I'm working across a half dozen different programming languages, multiple platforms, many endpoints and toolings, and operations and mathematics that I constantly have to refresh myself on. If I was doing copy/pasta template code all day in a tiny little niche I probably wouldn't find LLMs useful. But I don't, so I do.
> Being jaded and cynical about it doesn't change that reality.
Downvotetards need to understand this harder to break out of their funk. Criminally underrated comment, thank you and I wish I could give you more karma.
> It’s worth noting that you’ll have to sign in to your Microsoft account to use Rewrite, as it’s “powered by a cloud-based service that requires authentication and authorization.”
They are truly data leeches trying to lap at any cut you might have for a bit of tasty data.
It’s pretty obvious after their data grab with the new Outlook. They destroyed the built-in app which I used all the time just to try to get all of my emails.
The New Outlook(tm) is an Edge Webview to the cloud.
In the stupidest twist of fate, you cannot open Outlook offline, at all. There is no concept of "offline" in the New Outlook. I assume this is Microsoft forcing away the issues of the past of 50+GB OST files by making Outlook a glorified webmail client instead.
Oh, and Microsoft Teams? You can open that offline and it's got a full cached experience. Innovation at its finest!
This is just so sad. It's not bad, it's to be such a great application and something which had endured the whole Microsoft enshittification. But I guess even that falls.
Soon there won't be any safe place on Windows to copy and paste some random text/password/key that I don't want to get uploaded to some cloud service, or screenshotted and "analysed".
Even then you would need do that everytime because it could change at anytime or after updates. There is also nothing stopping them from detecting a MITM and acting differently.
How would you go about decrypting that traffic? I guess some kind of proxy, but I would assume that Windows would use certificate pinning so as to avoid MITM attacks. Another option?
"Introducing Windows 365 Universal Clipboard, ensure all of your random text/password/keys get uploaded to a cloud service, and your clipboard history preserved forever. Search your Windows clipboard history from your PC, phone or even television, anytime"
Doesn't Apple already do some of this with continuity clipboard? I can cmd-c text and files on my Mac and paste on any device signed into the same Apple account - I would assume this data is passing through iCloud.
Why do writing tools have to be a feature of every app instead of an operating-system wide tool like Apple Intelligence? Makes no sense whatsoever to just "sprinkle AI everywhere" except for specialized use cases.
You seem to be of the understanding that Microsoft has some sort of unified strategy that tries to optimize for user experience. Might have been true in the beginning of the company's life, but today? They've been using the "spraying and praying" strategy for the last 2 decades at least and this is just another way of applying said strategy.
I bet there's there's a poorly thought out directive from the company bigwigs to integrate AI into all Microsoft products, and middle management is gobbling that up to get ahead in the rat race.
Visual Studio and VSCode have also become infested with little Copilot icons.
I've yet to have a good experience with copilot in Visual Studio
It still surprises me to have shows stopping bugs with it, In THE first party IDE, in Windows, using pure .NET and other microsoft tooling
I couldn't describe a more perfectly vacuum'ed spherical cow, and still, copilot dies randomly even after they have acknowledged the problem and made some fixes
Visual Studio unfortunately still heavily uses legacy .NET Framework, only some parts of instrumentation specifically for C# were moved out of process and run on top of modern runtime. It's a pretty good IDE, much better than the kind of experience you get elsewhere, but I've moved to VSC + Rider since and only use VS occasionally on my "gaming pc" because it has a convenient community-made extension for getting quick .NET's compiler output.
As far as I’m aware, Windows doesn’t have anything quite like services on macOS and probably can’t because Windows apps, even those that are first-party, are built with a menagerie of different UI toolkits which means there are no universal hooks for something like services to use.
The reason macOS can do this is because a large majority of apps are either native AppKit or otherwise hook into the system text facilities (which is why text services work in text fields in Chrome and Firefox for example).
It's less effective at the OS wide level because the context is incredibly important. Notepad is used very differently from word processing and other applications. Even in the context of notepad- you're using that app very differently when working in python vs a readme.
AI is best integrated at the application level because the application developers are best suited to tune the algorithm and determine which pieces of context are most relevant.
I think this is where it will go inevitably, with only specialized AI tools being features baked into products you like.
A lot of our opinions on AI (like the one you just conveyed) stems from now nascent the design thinking around AI truly is. Like early internet, I suspect 10 years from now the UX of these tools will be much more mature. There'll likely always be a need for what ChatGPT has now, but these do feel like OS features.
> Makes no sense whatsoever to just "sprinkle AI everywhere" except for specialized use cases.
Microsoft really don't know what they're doing. They're trying to throw stuff at the wall and see what sticks without thinking of any externalities such as people getting pissed off and leaving to other platforms.
Despite how far its rotted, macOS is built on object-oriented APIs. Making Apple Intelligence work for text means adding it to NSText and whatever crappy UIKit objects came over with Catalyst/SwiftUI, not every single app individually bespoke.
Is the Windows Copilot not that? There has been annoying focus-killing long-form automatic autocomplete replacing the text popping in every kind of text area on the computer I use at work. Is this being done on each application?
Strange. Microsoft seems to struggle with the fact that they named it "Notepad" and some subset of users took this to heart and used it as a note pad, but due to backwards compatibility concerns it must never save rich text formatted files, else it could cause confusing data loss scenarios for users just trying to edit config files. Hence, the odd combination of LLM features added to a text editor that will never support rich text or rendered markdown.
Seriously, no one writes with Notepad. It's just for quick copy-paste text manipulation. That's it. No need for gen-AI.
I used Notepad to make simple HTML pages back in 1995 - maybe, two dozen lines of "<h1>" and "<blink>". That's the most text I've ever written in Notepad.
I write with Notepad. No distractions, no temptation to add anything but text.
I write musings, short stories, song lyrics, etc. and keep an archive of all these text files to reflect on later. Sometimes I write notes and letters to my future self.
It's basically autocomplete, spellcheck and refactoring, just based on a new technical approach. Those are not new or unusual features for a text editor.
On the other - the whole appeal of notepad was that it was a barebones text editor with none of that fluff (aka - it's a text editor, not a word processor or IDE).
MS has a large number of alternatives for the folks who wanted them.
When I opened Notepad - it's explicitly because I don't want the machine trying to tell me what I entered is wrong, or fix it. I just want a big dumb textbox for my file.
---
Basically - If you wanted those features you're looking for MS Word or Wordpad.
But MS discontinued Wordpad, and now they seem intent on trying to turn Notepad into Wordpad 2.0.
My prediction is this will not work well, since it directly competes with Office, and is not what the legacy users of notepad want.
But hey... a text editor is an easy place to shove text based AI, and cool shiny new thing of the year means some exec can claim to be shoving novel solutions into prod and bump up AI usage.
I just think people want a nice zen experience in notepad without the buzz of tech and AI swarming them. Kind of like choosing pen and paper instead of opening up Word. If I want a text editor for "all of the above capabilities" it won't be notepad.
Notes is a very different app to Notepad though. Nobody uses the Notes app to edit plain text files, which is the sole use case of Notepad. Notes is for writing, well, notes. Notes also has a lot of other editing modes and features like drawing with the apple pencil, scanning documents, cross device syncing, etc. As far as I’m aware Notepad can only edit plain text files.
It's unfortunate that Apple Notes doesn't handle plain text. I'd use it like this if possible.
Anyway, I use TextEdit in plain text and autocomplete, autocorrect and spellcheck all work just fine, as they work in every text box in macOS. That Windows' Notepad got some of that just in 2024[1] is bonkers…
I used Notes to edit a text file once and learned a very important lesson; It changes all your double-quotes to fancy Unicode double-quotes. They may look prettier, but they completely break the code you are working on.
I wonder when the environmentalists will realize what is going on right now, and start to protest AI usage in general. I was already wondering the same when the open source community started to CI every damn PR and commit, but I guess I was too optimistic with that one.
I don't want to sound skeptical, but this is what Crypto people used to say when it was very new.
It was supposedly worth all the power expenditure, because changing the world needed energy. Now we see where we are.
I'm inside this "newfangled AI thing". There are groups which create value, but they create value for everybody. The humans and the nature in general, and they use AI for scientific ends. Medical image processing, ecosystem monitoring, etc. etc.
Letting bots loose on the internet, letting them consume what they say and making them answer "Sauce is a food taste enhancer, and dressing is used to keep wounds clean while allowing them to heal. A standard serving of a dressing is two spoons".
To be fair, newer research is demonstrating that smaller more power efficient models with the same performance are possible, so the hope is that these giant LLMs are just a stepping stone to a less energy hungry place. In contrast, proof of work fundamentally needs more energy then bigger the network gets. It's no guarantee but we can at least see some hope that as energy impact drops and increasing value is found that 'AI' will cross the threshold of being worth the energy.
Edit: although yes I do agree that the 'value' part is tricky. If internet spam can generate more 'value' for some people than doing science, then when intelligence is cheap we are in for a rough time.
To be clear, I'm not against AI or LLM as a technology in general. What I'm against is the unethical way how these LLMs trained and how people are dismissive of the damage they're doing and saying "we're doing something amazing, we need no permission".
Also, I'm very aware that there are many smaller models in production which can run real-time with negligible power and memory requirements (i.e. see human/animal detection models in mirrorless cameras, esp. Sony and Fuji).
However, to be honest I didn't see the same research on LLMs yet. Can you share if you have any, because I'd be glad to read them.
Lastly, I'm aware that AI is not something only covers object detection, NLP, etc. You can create very useful and light AI systems for many problems, but how LLMs pumped with that unstopping hype machine bothers me a lot.
I disagree. Crypto people kept suggesting that crypto was a solution to an X problem while ignoring that a database was a better solution the problem.
I’ve yet to hear any good use cases for crypto, and I’ve been asking for years on here. Meanwhile there are a bunch of AI tools out there that are working and helping.
AI is a gigantic landscape with tons of different applications to different problems, and there are many solutions which work for a given problem.
However, if we narrow what AI is to LLMs, we have a stochastic parrot which needs to be fed the world literally to enable it to create semi-coherent sentences about something being asked. More importantly, what that parrot says doesn't have to be true, it can't be guaranteed to be true, and can't be verified about its accuracy about its slop.
And you spend gigawatts of power just to train this thing which selects and prints words based on probability and some randomness.
I think it's premature to be integrating LLMs into operating systems. That said I think they're very valuable, and the training is fundamental research. I feel like complaining about the resources used to train new models is a bit like complaining about the resources used to build experimental fusion reactors or particle accelerators. The fact that we're seeing direct applications is a bonus, but it's still more like fundamental research than anything.
People are spending all that money training because they are trying to fix the problems you're complaining about, and this includes fixing the power consumption problem. If we can create 3B parameter models that have capabilities on par with today's 405B parameter models, that's worth spending a lot of energy training. But nobody knows what is possible, so they have to try. I feel like you're basically arguing nobody should try because you don't believe they will ever improve, but that seems contradicted by the general trajectory of how things have been working the past decade. More resources spent on training means more efficient and useful models.
You've moved the goalposts from AI to LLM's. Fair enough, we've been doing AI since the 50's, and this is the second AI boom in a decade.
Those "stochastic parrots" have still proven that they are immensely useful. You might not personally find value out of coding assistants, but many many people do (as an example). People are (allegedly) turning to LLM's rather than StackOverflow for help [0]. They work well for boilerplate where you're an SME and able to validate the output - I can review 10x the amount of code I can write for example. They work (remarkably) well for summarising input text. An example - I semi occasionally (3-4x per year) have to deal with a few hundred GB of audio files that need cleanup. The cleanup tasks are "run FFMPEG with parameters", except I can not ever remember the parameters (they're different for different things). I can: read https://ffmpeg.org/ffmpeg.html or I can ask ChatGPT to write a script to clip the silence and add a 0.5 second intro fade to every file in a specified folder, and the entire task is done before I've even thought about it. I get to focus on what I want to, rather than munging data around.
If you expand your definition from LLMs to Transformers, then you get Whisper as a stand out example of something awesome. There's definitely negatives, but things like Diffusion are being used outside of image generation for drug discovery. We're not going to yolo AI generated drugs into human testing, but we can save an awful lot of screwing around to find something viable.
> And you spend gigawatts of power just to train this thing which selects and prints words based on probability and some randomness.
> That doesn't solve any problems.
I disagree, it does solve problems. A very fair question to ask is "is it worth the cost" and I would agree that it's not worth the cost. That doesn't mean it doesn't solve real problems.
> Crypto people kept suggesting that crypto was a solution to an X problem while ignoring that a database was a better solution the problem.
Only if you ignore the fact that a database (in traditional sense) doesn't solve the problem of decentralized peer to peer payments, which is the key differentiator of cryptocurrencies.
> I’ve yet to hear any good use cases for crypto, and I’ve been asking for years on here.
If you've been asking for years, I'm sure that someone, at some point, has told you about crypto's censorship resistance and international payments in places that are poorly served by the banking system for a variety of reasons.
Would you like to hear more or have you already dismissed these as "not good use cases"? It would be nice to differentiate between use cases that don't apply to you personally, and use cases that don't apply to anyone.
> Only if you ignore the fact that a database (in traditional sense) doesn't solve the problem of decentralized peer to peer payments, which is the key differentiator of cryptocurrencies.
And the point that the useful cases cede to make a useful product. The thing that is a _feature_ of a cryptocurrency is why people don't use it. I've had this debate dozens of times on here.
> If you've been asking for years, I'm sure that someone, at some point, has told you about crypto's censorship resistance and international payments in places that are poorly served by the banking system for a variety of reasons.
You know what else solves that? Cash and Western Union. And it has done for a long, long time.
> The thing that is a _feature_ of a cryptocurrency is why people don't use it.
People do, in fact, use them. Is it a popular payment method in western countries? No, but do some people use it? Yes, they do.
For privileged people, decentralization is usually a serious flaw. For others, it's an extremely important characteristic that lets them transact at all. The world isn't black and white, and people have use cases that are different from yours.
You're being self-centered, and that's okay, but perhaps you should factor it into your mental model before making sweeping statements in front of a global audience.
> You know what else solves that? Cash and Western Union. And it has done for a long, long time.
Not nearly as well, or there wouldn't be anyone using cryptocurrencies for that purpose.
You could make identical boring, bad-faith arguments about AI products. I think 99.99% of all "AI" products available today are completely useless - to me - but I don't go around proclaiming that all of AI is completely useless, and that all of its problem areas are better solved by statistics and "if" statements.
Don't mistake your own privilege, ignorance, and lack of imagination with the lack of real-world applications.
> You're being self-centered, and that's okay, but perhaps you should factor it into your mental model before making sweeping statements in front of a global audience.
> but I don't go around proclaiming that all of AI is completely useless, and that all of its problem areas are better solved by statistics and "if" statements.
> Don't mistake your own privilege, ignorance, and lack of imagination with the lack of real-world applications.
If you can't make your point without making sniping attacks about my character, then this isn't a conversation I want to continue having.
Privileged person is anyone living in a western country who hasn't had to deal with censorship. I consider myself to be a privileged person in that regard. That's not an attack on anyone's character.
> You're being self-centered
That's anyone who fails to consider use cases other than their own. I wasn't speaking to your character, It was a description of your reply, not your character, because it contained sweeping statements that only apply to certain groups of people.
> but I don't go around proclaiming that all of AI is completely useless, and that all of its problem areas are better solved by statistics and "if" statements.
That's not an attack on anyone?
> Don't mistake your own privilege, ignorance, and lack of imagination with the lack of real-world applications.
I've explained that privilege isn't an attack on anyone's character. As for the rest, sorry, but which words am I supposed to use when someone denies that a problem is real (which fair enough, I'll elaborate), later admits that there are other services that solve the same problem, but they still want to claim that there are no problems that the obscure product is solving, despite that product having real-world users who are using it for that exact problem?
In terms of generative AI, for general use cases, Open AI reported having 11 million paid subscribers last quarter. People clearing paying for Adobe Firefly and Midjourney access. That's already a ton more people finding it useful in day to day life than Crypto ever had.
It is certainly reasonable to suspect that the scale of investments (in trillions of dollars) don't match the scale of the opportunity. But it's a bit silly to pretend that no one is getting any value out of this.
At the end of the day, data centers are 2% of energy use, according to the IEA. That's trending up, but even in couple of years, data center stuff is mostly going to be typical cloud stuff, then crypto, and then a fraction for AI.
I never heard of any reasonable uses for Crypto or Blockchain in general. A lot of people tried to sell us various things at the time but it was very obvious that it had no real value.
AI is already implemented into businesses in various ways. Even if it’s not done so official you still have loads of employees pouring company secrets into chatGPT and Claude because they work.
Manga Library Z, a manga archiving site that distributed old and out-of-print manga for free has been forced to close down due to all major credit card companies refusing to provide payment services. If some hypothetical widespread decentralized payment system can prevent scenarios like this one from happening, then it would be worth the "enormous waste of resources". These days, you're essentially relegated to a non-person if card companies stop allowing you to use their services.
Windows Update is another big one, the way it makes every Windows computer spin the fans like crazy on a regular basis. Probably room for improvement there.
maybe they could just make the software not be completely shit instead ?
why does Windows take 2 minutes to decide what updates to install, then 45 minutes to install them when Debian on the same machine can do both in under 30 seconds?
Coming soon: "To ensure users have the best experience, we will require that AI features be active by default, which requires a continuous internet connection, a logged in Microsoft account, and enabled telemetry."
I think it would be much easier and faster to just give the users one option. That way, they don't need to think about what they want and can focus on their work. (satire)
My guess would be that there's probably some correlation between lass choice and decision fatigue.
Whenever I go grocery shopping it always takes me quite a while to decide which rice is worth my hard earned money and which isn't, especially because some rice makes my tiny rice cooker to start foaming all over.
"Now that you are having the best experience, to ensure safety we will require acknowledgement that we might monitor what you type and may have to ability to disable your required account if we detect unsafe words or searches. This can mean that you will not be able to log back in to your computer. If you experience this issue simply log into your account and contact our automated support system."
The addition of tabs to Notepad has made me stop using it entirely. My entire workflow around it was having separate windows for each document that I could move and close freely. Having an all-in-one-place Notepad filled with the ghosts of API tokens past is precisely the opposite of what I needed.
I think it takes maybe 3 clicks to set up "old style notepad". I don't understand the harsh criticism? I don't use it myself since I have a sublime license, and st3 opens up instantly, but I don't think the new changes to notepad are the end of the world.
I don’t understand why anything I said is a “harsh criticism” or “the end of the world.” I don’t actually care about Notepad! It might have taken 3 clicks to set it up, but it took zero clicks to say “this is a change I didn’t want and I don’t care enough to learn how to fix it, I’ll just use something else.”
Yeah, not everything needs an LLM tacked on. Notepad is a lesson in tool minimalism; it serves a lot of use cases precisely because it has a small feature set.
I'm not sure how they built this thing, but it makes sense to me on the face of it: Notepad should be the thinnest wrapper possible around a simple textbox, and the generated thing they're showing should be built into the widget itself so it's exposed to all apps in the OS (much like macOS exposes emacs/readline-style shortcuts everywhere that uses a default textbox).
This seems to be a common trend everywhere. Instead of improving core functionality they add the latest buzzword tech, no matter it make sense or not. I am user of GaiaGPS to plan outdoors trips. Same there: the app is good but has some problems. Instead of fixing the problems they added social functionality which nobody really wants.
Couldn't disagree more. Notepad is a place to dump snippets of unformatted text, for a temporary duration unless I explicity save the file. I already didn't like that they added the tabs feature and autosave recently.
If I want to do writing I'll use one of the 6 tools on my PC more suited to that task.
As a counterpoint, I'm on macOS where their latest AI writing tools are now implemented system wide. I also use TextEdit (approximately equivalent to Windows Notepad) and the Stickies app for similar text dumping ground behavior and yet, having the AI writing tools available on those two apps is incredibly useful. I often don't use it at all, but there are things I want to run through it and not having to move the text yet again to another app is nice.
It's _also_ useful in my apps dedicated to writing and even the text areas of browsers. I think it's all about implementation though, Apple's writing tools are quietly buried in the context menu for most text inputs. Microsoft has a tendency to be pushy and in your face about their latest AI offerings like shoving it into the Start Menu, or making it a prominent and visible element of their UI (Copilot in VSCode, even when you're not a subscriber) and the Verge's screenshot isn't enough for me to judge this by.
Word is the perfect place for it? Or OneNote. This would be like if Apple added AI rewrite to TextEdit instead of Apple Notes. Notepad's only job is to open text files as fast as possible.
> And clearly its a useful too in a writing app that users would like
I don't believe it's useful. I don't think Notepad is a "writing app." And I'm fairly certain not a single user in the history of ever has asked for this.
Nobody got promoted for improving core functionality.
It depends on the company.
If you work for a hyper-scale tech company that only cares about money money money, then yeah -- nobody's getting promoted for improving core functionality.
But I've worked at several companies where that sort of thing is not only rewarded, but celebrated. One was a factory. Another was healthcare. Tech is the aberation, but on HN we pretend that it's normal and good.
People get promoted for improving key metrics. Like Revenue or Retention or Installs or whatever. Metrics can be abused and gamed ... but you can do a better job aligning them to improvement than the de-facto metrics of "number buzzwords in a new shiny toy"
This is your tried and true. C level read a magazine, mandatory implementation order, engineer ticks the checkbox and waits for hype to pass so they can remove it.
A scary version of this is driven home to me when I go to Washington DC and see all of the very expensive billboards at commuter stations near the Pentagon advertising fighter jets and other military equipment.
It scares me every time because they wouldn't be splashing out the big bucks for those billboards if they weren't effective, and I absolutely don't want the military (or any other entity engaging in major expenditures) to be making those decisions based on billboards.
Maybe taking a step back here, what core functionality is missing from notepad? I see this as a fairly feature complete tool for a core set of behaviors already.
I guess it's a good signal for when a niche is opening for someone to develop a better product. That's probably a reason why competitors like onX have sprung up recently. (Although I haven't used onX so I don't know if it's actually better or not.)
Also with Gaia it feels like very single year they just have to tweak and change the interface to the point where nothing is easy or intuitive anymore. It's like designers need to justify their jobs so they just keep making change on top of change and departing from what made sense, just to make changes, just to be doing something.
I have tried OnX and Caltopo and I think for Gaia is still the best. If they just fixed their folder management. It makes me angry every time I have to use it.
Why not just spend the one time cost of the approximately four seconds it takes to remove Copilot and save yourself from feeling drained if it's that upsetting? This is right below the level of effort involved in changing the theme and setting the font in your IDE.
The chips are improving. They are adding dedicated AI co-processors/cores. Every major chip developer is moving towards providing functionality that supports running AI models locally on chip. Eventually all of these use cases will be run locally, fast, and power efficiently w/o network hops. Waiting until that happens to develop these features puts you ages behind everyone else. It's just another iteration of the thin-client to fat client product cycle, and for AI, it's very early.
It seems not enough people are using AI features and therefore it is being put everywhere so that people can't help but use AI features.
Instead of realising that people don't want these features and actively disable or avoid them, they've come to the conclusion that they're not being used because they're prominent enough, so they are being put everywhere so they cannot be missed.
Windows is no longer a Personal Computer Operating System that you can just use.
They spent all this money on training. Massive amounts of one time and recurring costs. They have no idea how to write it all down. They never planned that they might have to.
And people wonder why some of us were so insistent on not allowing this garbage pile of technologies to be open labeled 'AI.'
It's been pretty fascinating to watch what Microsofties have to put up with, having switched to Linux in 2007. Vista's half-assed release, Windows 8's tablet UI disaster, Windows 10's online account requirements, now all the AI crap they're shoving into Windows 11. Meanwhile I'm just over here using basically the same Arch Linux tools & XFCE environment for almost 20 years. Yes, I know, Linux isn't perfect, but it's maybe worth a look if you're fed up with using an OS that wants to exploit you instead of serve you. It's especially easy now that most things are web-based.
Arch has consistently proven itself to be the perfect machine for me. I go for months without a reboot even with daily package updates. 2000 line C window managers like DWM leave me more productive than windows ever has
I stopped using Windows and switched to Linux exclusively decades ago. But I don't get to choose what OS I use at work, so what Windows does still deeply affects me.
"Just switch" is great for home machines, but for most, that's not an option at work.
Work is for work stuff. It's someone else computer, someone else wants and needs. I don't mind if they want to feed all their data to Big Tech (On my previous work computer the only truly personal piece of data I had was my profile picture and ssh keys for my GitHub account). I don't even do web searches on work computers. If Windows is slow, I'd just bring it up to my manager and the IT dept.
On my personal computers, that's a different story.
I totally agree, but I was really addressing just having to put up with the OS rather than privacy issues specifically. My employer can take whatever privacy risks they want.
I do get your point. But all of them are fine OSes, if you just want to run some software (which is the majority of the jobs). What macOS and Windows don’t allow is true customization, aka aligning it to your personal taste. That’s not great for a personal computer, but fine for a work one. If the company don’t want to address inefficiency, that’s on them.
I like how currently on LinkedIn there’s a feature to use AI writing assistance to help land your next job. Like if I can’t write how am I ever going to work a job
I recently used ChatGPT canvas to improve a friend's resume. I had managed him in the past, and he is great to work with. However, from his resume you would never know it. I did 3 hours worth of editing and improvement in 5 minutes. I gave rambling descriptions of what he had done when working for me and why it was impressive, and GPT translated this into resume speak almost instantly. I then gave some vague suggestions for improvement, made a couple minor edits to buff off the AI veneer, and viola, he had a professional resume that will help him do great work in his next position.
Seriously, a use case where LLMs really shine: bullshit padding. It's quite amazing to be able to turn ramble into coherent text. The less obvious tax is that if you don't put effort you will never be able to learn how to write and that will most likely affect thinking too. Writing will remain a valuable tool for improving thinking though, but likely for a lot fewer people.
I suspect we'll start to see people asking for more raw bullet-points for certain communications, because all the text they used to use to guess engagement/thought/investment/care is debased from easy counterfeiting.
In some cases that'll remove something that was always a questionable waste of time, and in other cases it'll represent a real loss of information.
Did you consider that the AI screeners would have liked the AI veneer? That CV will be "read" more times by a machine than it will be by humans. Someone should publish an RFC for a machine-to-machine resume spec so we can ditch the inconvenience of using human language as the transport
Unless and until I see a clear privacy police analyzed by a lawyer, stating categorically that this doesn't happen, I think it's wise to assume that you are correct here.
While it's possible they meant something nefarious, I think the goal was probably just to prevent confusion around shared computers. Such confusion might seem unlikely, but when your market is as big as Microsoft's even unlikely things happen all the time.
Microsoft's "CoPilot PCs" scheme places a lot of emphasis on the hardware being powerful enough for locally-processed "AI" features and isn't about offloading everything (I suspect because they're starting to acknowledge that not everyone has a highly-reliable multi-megabit Internet connection).
It a result of a rather sad trend, which I think started with Google. Rather than going out and doing market research you just throw things against the wall, measure and see if people are using it at any significant rate.
The difference between Google and Microsoft is that Google have no problems just killing of things that doesn't perform to their standard (which is bad in it's own way). With Microsoft backwards compatibility is everything, so once something is in Windows, it says around for a very, very, long time.
AI assistance in writing isn't a bad idea, but maybe not in Notepad. I know that this isn't they way modern Microsoft wants to do things, but exposing an API that would allow 3rd. party vendors to AI support in Windows seems like a more sensible approach. Except they'd probably have to make it accessible from Javascript to make anyone use it.
Perhaps trying random things uncovers more big opportunities than market research does. Market research tells you to breed faster horses; random things converge to combustion engines.
According to Wikipedia [0], the development of the car concept went from 1649 to 1881. That's not exactly throwing random things and seeing what sticks. It may have not been systematic, but with the wealth of information and resources we have today, we can and should do better.
I hate that Win11 Notepad has so many tabs open even after you close the app.
Many times, I just paste my copied text in Notepad to strip the formatting + special characters and close it after re-copying the data. Pretty efficient.
Now that I open Notepad.. all my previous tabs are open asking me to close them one after the other (extra click on not so save the file) :@ so annoying
I'm not even complaining anymore. I'm just ceasing to use this Ai garbage and ceasing to deal with people who spruik it. And if you're wondering what "spruik" means, Google will give you an Ai summary which is subtly wrong and contradicted by the first actual search result it list directly below it.
A long time ago (in the 2000s) I was interviewed by a guy who was the maintainer for notepad. It was a part time job. Now I’m sure there’s a whole team around it with multiple devs, a designer, a product manager, a program manager, and now with AI, legal has got to be involved.
The value of notepad and paint to me are that they are simple and reasonably minimalistic. Adding features makes those sorts of programs less useful to me.
Also, I can't trust any software that incorporates AI because I can't trust that my usage of that software won't be used to train AI.
So this move means that I won't be using these tools anymore. Not that it matters to anyone aside from myself, of course. It's also not a huge deal to me -- I'll just start using different tools that still meet my needs.
Genuine question, what are these companies hoping for in return of giving this for free? Not just giving it, in fact pushing it hard. Do they want users to get addicted now and start charging them later? Do they want to increase Windows retention? Will they push subtle ads?
I think the basic text widget of all the toolkit support text attributes (color, styles,…) and as the solutions for syntax highlighting and other stuff existed already, it was not a big stretch to add them instead of having two editors (the first user was the programmer)
I don't blame Satya, I blame the foaming at the mouth AI-frenzied investor class. If these folks thought a second about this tech and tried to use it in a number of domains, they would not be clamoring to AI all the things.
If you listen to Satya Nadella various talks/interviews you will realize that he think AI is the next major platform and we know MS lost the last major platform i.e mobile. So in his vision he wants MS to be the major player in this next platform shift and the way they are doing it is by putting AI in every nook and corner of their OS and apps. The only issue is that AI (in its current form) is NOT a platform but more of a "user interface based on natural language".
Well, at least now it will have a semi-valid reason to connect to the internet. It is still bonkers, but a different kind of more acceptable kind of bonkers.
I just wish Microsoft had a simpler mindset for their OS. Simple, privacy-first, consumer-first defaults and optional upgrades to more enhanced tools via their App Store.
Imagine if instead of Windows Recall being installed and available automatically on machines, they just added Recall as an optional downloadable add-on via the App Store... I don't think it would have received nearly as much backlash.
I don't think creating a separate tier of enhanced OS upgrades would benefit Ubuntu. Ubuntu isn't connected to a multi-billion dollar corporation. Canonical has no resources to offer a free AI-powered Notepad editor. Microsoft is connected to OpenAI, Microsoft has AI-hardware partners, and Microsoft has the in-house resources to create new drivers for new hardware/software compatibility issues.
There will always be valid reasons to use Windows over Ubuntu.
Does this mean its going to bloated and slow to open. I want notepad to be absolutely instant. Even calculator is getting slower to open now. I want these simple apps to be fast af.
When "social" was a hot new thing, Google suddenly wanted to be Facebook, and started making all of their products part of Google+, whether it made sense or not.
Google+ required having a public profile with a real name. Gmail/calendar/docs users who were tricked into activating plus, but didn't comply, got banned or forced to make a real name profile. Merging of Hangouts, Orkut, Blogger, YouTube comments, and Play Store reviews into the same "social network" made no sense. It was a blatant move to inflate user numbers, at cost of annoying users of the other services, who didn't sign up for a me-too Facebook. That was the first time when Google became uncool.
It kind of is for individuals. They don't bother punishing you too hard for pirating it anymore. I can't change my wallpaper unless I go legit? Oooo, I'm so scared!
Microsoft remembers Bill Gates's mantra that piracy at the individual level actually gets you an audience who will then sign on the dotted line once the big-money deals -- enterprise license contracts -- need to be made.
>It’s worth noting that you’ll have to sign in to your Microsoft account to use Rewrite, as it’s “powered by a cloud-based service that requires authentication and authorization.”
What?!
What about AI PC, NPUs in e.g Lunar Lake and in general AI@Edge?
I do not know much about the internal workings of Windows, but why is this not implemented as a system wide action? Are they really implementing "AI Tools" in every application that uses text... individually?
If this was in the context menu, along with Cut, Copy and Paste, you would be getting AI integration for free on every application that renders a text box widget.
I mean... spell check is implemented like that right? They seem to be pitching this as a way to correct or improve text so why not build on that?
I'm not even a fan of this sort of generative AI "type whatever" tool, or I at least find it to normally be inadequate. But if they're going to do this AI pivot in Windows anyway, the choice of implementation here seems short sighted.
people already forgot that microsoft said years ago that they will make windows into a service and there will be no new version, just constant updates. this is what it looks like. ZERO privacy. you'll own nothing and be happy.
Wth is the point of this? Are they trying to get data from every source they can? This can mean one thing: they want to get eid of openAI in years to come so they are collecting everything they can.
