If you read the article, they mention that they’ve reported the bug in previous versions of the beta and it has still not been fixed in the latest version. They’re cautioning their users against upgrading in two weeks when the release comes out of beta unless there is confirmation that the bug has been fixed.
Return To Office in early 2023 caused many people hired during the pandemic to uproot their family and move closer to their assigned office.
Some of these folks are now being told that the office they moved closer to is not a “Hub” for their organization and that they now need to Relocate To Hub.
Most engineering teams will not be colocated even after this relocation as there a multiple hubs.
There is a strong belief that Amazon will have a 5 days in office policy starting after the holidays and further Relocate To Team initiatives. The delay is to mitigate the risk of attrition affecting Peak and to get people to move before they’re told they need to be in the office 5 days (sunk cost).
A textbook lesson in how to boil a frog courtesy of McKinsey. Hopefully customers enjoy the taste of boiled frogs.
>There is a strong belief that Amazon will have a 5 days in office policy starting after the holidays
It's not just a belief and it's not just Amazon. Anyone who has been paying attention knows that "mandatory 3 days/wk RTO" will quickly morph to "recommended 5 days/wk RTO" which will then morph to "mandatory 5 days/wk RTO". The endgame is very much a return to status quo ante, i.e. full time in office; hybrid was always intended to be transitory despite the messaging.
I’m located in the US. Of the last 4 pieces of media I’ve bought on Amazon, one was a misprint that should have been destroyed, two were marked as for sale only in India, and one was an EU market copy. All were marked as new. I returned all of them and found the same product locally within three blocks of my apartment for the same price I paid on Amazon.
I’m surprised the big publishers (both print and gaming) such as Pearson and Nintendo aren’t putting more pressure on Amazon.
I started with an apple keyboard. I bought a Das in the early 2010s. When the das quickly ran into issues, I got an HHKB pro 2, which I’ve taken in my bag every day to every job I’ve had and I use at home[0]. Your comment just made me realize that USB hubs aren’t on every keyboard!
[0] the swappable USB cable is so handy! Just leave one plugged into each machine.
You realize that Ada and Objective C predate Object Pascal, and Erlang came out in the same year as it, right?
I would worry less about new ways of doing the same thing and more about client generation/retention. Plenty of companies are willing to pay top dollar for boring technology.
It wasn't a chronological order, of course...
But if before the languages were more or less field specific, now with each language you can build any kind of application and it became very difficult to make a choice or simply keep up with the daily updates.
I don’t think you understand how moat is used in this context. A moat isn’t impossible to cross, just harder to cross than a lawn. Also, I don’t think you can saw two features are the same unless the quality is the same. Two sms apps are not equal if one of them occasionally fails to send/receive messages.
I agree that postman has a moat. I think it’s important to say that every company has a moat in that they’re organized, located somewhere, and have customers. The size of the moat varies. We can treat a moat as a scalar value of dollars to cross.
Every company also has a valuation, either through public markets, private markets, or discounted cash flows.
A house cleaning company has a moat of trained cleaners, existing client generation process, goodwill of existing clients, etc. If the value of their discounted cash flows exceed the cost of crossing their moat, they are vulnerable.
The argument is that crossing postman’s moat would cost much less than their private valuation, and their future roadmap is unlikely to build a moat that is significantly harder to cross.
Does postman do anything curl doesn’t? When I was on onboarding at a job my pair started teaching me postman, which ended with me teaching them curl and them dropping postman. If you need to share queries, put them in version control!
It’s like saying web browser doesn’t do anything that curl doesn’t. It’s technically true but not really.
But it’s also technically incorrect. Even on request level, not speaking about all the things on top that Postman does, Postman also does grpc and graphql (edit: and websocket), which curl doesn’t.
Once you start adding all the features, it’s very hard to replace Postman with anything.
The advantage of Bruno (and the VS code plugin) that storing data in "just a file" is friendly to keeping a local copy, and also to sharing, reviewing and versioning with existing tools (git).
Not "weird" -- that's literally the operational model of VC. It is all about increasing the valuation of the invested company, as quickly as possible, and then selling it while the evaluation is high. Once a company gets VC investment, this is the only route that matters (unless the board somehow resists the dominance of the VC shareholders, which is extremely difficult to achieve).
I have a few qualms with this app:
1. For a Linux user, you can already build such a system yourself quite trivially by getting an FTP account, mounting it locally with curlftpfs, and then using SVN or CVS on the mounted filesystem. From Windows or Mac, this FTP account could be accessed through built-in software.
2. It doesn't actually replace a USB drive. Most people I know e-mail files to themselves or host them somewhere online to be able to perform presentations, but they still carry a USB drive in case there are connectivity problems. This does not solve the connectivity issue.
3. It does not seem very "viral" or income-generating. I know this is premature at this point, but without charging users for the service, is it reasonable to expect to make money off of this?
Your first point is almost a meme at this point - so many products can be written off as pointless with the statement:
> As a [type of tech person], [product] is completely unnecessary as you can easily build a similar system yourself by simply combining [tool A] with [tool B] and [tool C]. It's so simple!
I'm not disagreeing with your point, but it misses the point that often the benefit of using a particular product is that you don't have to build it yourself, not to mention the other benefits you could get from using the exact same system as somebody else (e.g. sharing Postman collections).
Classic HN response–"This is trivial to build"/"I could build this in a weekend".
Setting valuation aside, clearly there is some non-negative value prop for some set of users who don't want to build everything themselves (as evidenced by the fact that they have a non-trivial number of paying users).
Yes! It's supports functional BDD-style tests, "mock servers" (really "fake" servers), a walled garden for saving and sharing data (to frustrate engineering managers and ops folks), perf tests (soon? now?), contract testing, and more. Did I mention that it's a GUI tool, so it must be "easier?"
They're trying to become a one-stop-shop for dumbing down web dev. Ideally, inside the walled garden. That has a certain appeal to management.
The contract testing is something I didn’t know about, I’ll have to look into it!
We have lots of unit, integration, and live canary testing, the postman usage was only during oncall/operations, so we weren’t using any of those neat features.
I like that they let me run same set of APIs against different environments.
Also I can stitch together different APIs by taking response from one API and pass it as request to other APIs. For instance I invoke login API, grab session ID from the response and pass it to other APIs which are behind login.
Of course one could do all of this in curl but it’ll be cumbersome. And yes, if all you want to do is invoke a single API then Postman is definitely an overkill.
It might, but I think the appeal of Postman is how straightforward it is to play with APIs across basically anything in the HTTP stack, with a nice GUI. Also, I thought Postman also had the ability to export your queries to curl if you want to need to use it outside Postman.
It does, but the support is not great. There’s active work to make it better. We don’t use WebSockets so that might be why my colleagues didn’t have too much attachment to postman. We have a list of bash functions we import into our zshrc/bashrc that just wrap around curl with some parts pre-filled and automatically grabbed creds from ENV as needed. It’s a better workflow since you don’t need to switch applications and you can chain calls together with some glue.
