Hacker News new | past | comments | ask | show | jobs | submit login
When operating "at web scale", users often simply don't have "real names" (plus.google.com)
155 points by saurik on July 24, 2011 | hide | past | favorite | 112 comments



This title is 5 words longer than it could be, because people don't have real names.

Apologies in advance for the self-citation:

http://www.kalzumeus.com/2010/06/17/falsehoods-programmers-b...

My name+ is confusing gibberish where I live. It is highly likely likely that when I have children, they will go through life having multiple sets of gibberish so that they can pick the non-gibberish option when asked "What is your real name?"

Does your startup really want to get into the is-that-gibberish-gibberish-or-am-I-just-ignorant-of-the-way-they-do-things-in-weird-places-like-the-United-States adjudication business? If so, try validating names. You will have loads of fun.

+ : Well, a name I go by, at any rate. There's at least eight different ways to write my "real name" correctly, not even counting nicknames, online handles, or the like.


FB bans my real name because one of it is deemed offensive.


I've heard that they have a streamlined process to avoid false positives by sending them additional documentation.


Well, I know people who IRL insist on being called by made-up names, mostly goths, you know, something like Darkraven Bloodmisery. Whatever they called themselves in Livejournal. Very few of them follow through tho' and fill in a deed poll form! For me that's the acid test: whatever it says on your passport is your real name. And government IT - the least efficient kind! - has managed to solve this. Why can't the mighty Google?


And government IT - the least efficient kind! - has managed to solve this.

Governments essentially avoid this problem, and they have for generations, because they realized that names are intractably hard. This is why they give people numbers instead.

Here's my name written in my passport (US): "MC KENZIE PATRICK JOHNATHAN" The Passport Control Center is the only organization which thinks I have a space in my name.

If you go to Cook County and grab a copy of my birth certificate, it says "Patrick Jonathan McKenzie". I was a poor speller as a kid and thought Johnathan had two Hs until I was a teenager. Whoopsie. Now it does because it's just too much hassle to go get it changed now.

The Illinois DMV thinks I'm Patrick McKenzie. Ogaki City Hall thinks I'm McKenzie Patrick Johnathan in one part of the office, ミッケンジーパトリック in another part, パトリックミッケンジー in another, and ミッケンジーP in a fourth part. The national tax agency said, direct quote, "Sorry about our software being unable to accept your name. On the bright side, we really couldn't care less, as long as you pay your taxes on time."

Well, I know people who IRL insist on being called by made-up names, mostly goths, you know, something like Darkraven Bloodmisery.

I know a lot of people IRL who insist on being called by made-up names. I know Japanese and Chinese people who have "use names", because some white folks think Hideyuki is a mouthful and couldn't guess that he's a man. I know neighborhoods in Chicago where Irish Catholics are so thick on the ground that you'd better come up with a nickname if you're a Patrick or a Mary because otherwise half the class will have a hash collision with you. (I was Patrick Johnathan -- the other Patrick Jonathan beat me at rock paper scissors and claimed the coveted PJ.)

I also know some goths who use made up names, including a Robert who -- cheeky blighter -- insists he is really a Bob.

Also, just a self-preservation tip: don't ever repeat that comment near a liberal arts faculty if you value your sanity. They will be on you like vegetarian tigers/esses/etc on a fresh, juicy bit of fair trade tofu.


For a quick hint -- to attempt to establish the gender of a name from another culture -- Image Search is fantastic. I'm one of these white folk with no idea whether Hideyuki is generally a male or a female name -- but a Image Search results page full of male faces is pretty conclusive.

General caveats apply about edge cases (and Gender is specifically complex issue that maybe this advice is useless)


And yet, in Germany or Austria, 'Maria' can be a man's name.


Only as a second middle name. (Klaus Maria Brandauer)


Government IT.

I have legally had a one-word name for decades. SS card. Driver license. On credit cards I fake a first initial, same with other DB-enslaved contexts.

Some time after the DHS was invented, SS changed their database systems. Where the SS used to list me as "myname," now I'm listed as "#unk myname."

Since my SS record no longer matched my driver license, when I went to renew my license I was refused, because the DHS tells the states that DL must match SS. That took weeks to sort out, but I did finally get my DL properly renewed, after they found the employee that knew how the DB system works.

Government IT.


I believe that a point that patio11 is arguing that all names are made up. Some names are given certain legal uses over others, but this is not universal. Google is an international organization, that is attempting to form a social network application, so it cannot assume, well, anything about its users cultures and how those users use names. This is a huge problem, because the first thing that a social networking application has to do is: "Make people feel welcome". Telling them that the name that everyone uses to identify them isn't a 'real name' doesn't help to do that.

Also, I know a fair number of people, who, for a variety of reasons, have their names in transition. These new names are not yet their official names, and they may change what they ultimately decide on. It's _really_ not the place of Google to effectively say "Make up your damn mind".


I believe that a point that patio11 is arguing that all names are made up.

Well, there's a continuum of "made up", going all the way from Darkbane Bloodlovin' to our idealized gold standard of "Whatever two inexpert people who have known you for less than a few hours decided would be the best thing for the entire world to know you as for the rest of your life."

P.S. Got to love that I'm referred to as "patio11" in that comment. (I don't know where "My friend in middle school was from Puerto Rico and he just couldn't manage -ck most days so 'Patio' it was" fits on the continuum, but it stuck so hard that even my siblings routinely use it now, so I won't complain about it.)


For me that's the acid test: whatever it says on your passport is your real name.

Today I learned that I don't have a real name.

More seriously, though, most people I know IRL insist on not going by the name on their passport. If I had one, my name would be "Christopher", but I insist on "Chris" enough that I will not make my name "Christopher" on Google.

For a less silly example, though, most people I know whose names are not derived from their full legal names are some form of non-cisgendered people who simply don't identify with their given names. And, in the US at least, for most purposes it is simply not necessary to go the court-decreed route (we don't have a strict equivalence of a deed poll form for name changes as far as I know) and bringing court and public record into things can be unnecessary trouble until they fall out of those "most purposes".

Of course, these are only a couple of reasons why one might wish not to use their passport names online, but the main point is: it's rarely as simple as you think.


> "And government IT - the least efficient kind! - has managed to solve this. Why can't the mighty Google?"

Because government has really one use for your name: the proper identification of a person for the purposes of the provision and execution of government responsibilities and services.

Google has a larger mandate, one where enforcing "official passport names" is not only inconvenient, but would also drive away adoption.

Not to mention, I have an English name, and a Chinese name. Only one of these appears on my passport, but depending on who I am interacting with I would very much like to (justifiably so) use my "unofficial" name. What is wrong with this? This is actually somewhat annoying for Facebook - I have relatively in Asia who don't speak English at all, and I'd like for them to be able to find me on FB... but there is only one canonical name on an account.

Google is developing a social system - a social system by the very nature of human identity cannot just have "passport names".


>> Google is developing a social system

As I've said before, Facebook beat MySpace because it uses "real names". It's a lot easier to discover people you know, and that drives adoption. Google+ has to copy them, if they want to win the masses over.

They might be better off letting you have several, privately linked identities. For example, I could have friends under my "wisty" account, and friends under my real name. But they are best waiting until the muggles have already created great name accounts, and hooked up with their high-school semi-friends.

Once google is on near-equal terms with facebook, they can allow handles as well as real names. You could just move your "internet friends" circle over to your handle, and leave your "school friends" circle on your real name account. But until they have everyone's real name, this would be a poor business decision.


I know a couple of LJ'ing goths who now have Google+. Thus I can see people with stated first names like "Persephone", "Ariadne" etc. The rule of thumb is that if someone says that's their name, that's what you call them. There are also some clearly pseudonymous accounts, e.g. last names like "jumpycat", "Purple" etc.

It's impolite to ask what their "real name" is. I've seen someone (call her X) get upset because someone pressed the point of finding out what X's mother used to call her (if that's a a "real name" in any meaningful sense). And it was pointless too, what could you do with that information?

There's lot's of inaccurate information there anyway, I have two "Jo"s in my friends list. On is male, one female. In both cases "Jo" is a contraction, in one case of a middle name. They also chose to be called "Jo".

I don't really see why Google+ needs to know or care. What people answer to is what people know to look for, so that's what should be shown.


"The rule of thumb is that if someone says that's their name, that's what you call them."

Yes! Insofar as it can be made simple, that's how simple it is. I can't fucking believe how hard it is for some people (and databases) to accept that part.


e.g. last names like "jumpycat", "Purple" etc.

I reminded of Catherina Fake, a Flickr co-founder.


As far as I know "Fake" is the surname that she was born with.


Yes that's my point. You have to be careful about "clearly pseudonymous accounts"


All right then. I'll give Tallulah von Strumpet-Hausen and Martin Soulstealer the benefit of the doubt ;)

if your point is "you don't actually know" then mine is "you don't need to even care".


Not even my passport has it right. My "real name" is ridiculously long and formal, and I refuse to use it. Luckily my shortened name passes. Unfortunately it collides with a famous futbol star, so you never know what will happen.


On which passport? I have two.


With different names on each?


That's pretty common for dual American/Israeli citizenship. Lots of Jews have two names: A Hebrew name and an English name.

Sometimes they are transliterations of each other, sometimes translations, and sometimes no relationship between them at all.


Also, transliterations can change. "Bruce" Lee and "Jet" Li have the same "last" (family) name. But China decided to replace the English-friendly Wade-Giles with the more efficient pinyin, so younger Chinese will transliterate their names differently.


For me that's the acid test: whatever it says on your passport is your real name.

Here in Ireland we can translate our name in Irish and use that. I did that once for my passport. So I've had 2 different names on my passport without going to the deed poll, getting married, etc. all legally. What would be my 'real name' then?

(For the record the 2 string have a levenshtein distance of 10, so they are dissimmilar)


To be fair, the US government has a bludgeon that little old Google does not have - the threat of force.


Having a Dutch surname like "Spring in 't Veld" (roughly translates to "Jump in th' Field") causes all kinds of problems on form entries and editors that add smart quotes.

- Many non-sanitized SQL queries fail,

- buggy URL or HTML parsers create code like:

  /spring-in/
  <meta content='Spring in 't Veld ...
- Pagetitles like: Spring in /'t Veld

- Added smart quotes "Spring in ’t Veld" from Word or Rich Text Editors cause problems with sorting and identity consolidation.

- Stripping the quote character trips a 2-letter requirement.

- etc.

It is unlikely that people will name their son

  Robert';) DROP TABLE Students;--
But you should at least prepare your db queries for a quote. :)

http://xkcd.com/327/


Your roughly translated HN name is also funny: Blue buttock gargle ;)

But yes. Programmers have a hard time using unicode. It's a shame not all programming languages use unicode.

I also think is has to do with bad research. It isn't hard to check different types of family names and naming order.

There are a lot of countries where they place the family name first followed by there personal name.

I'm still trying to figure the best way to store names. Maybe 2 fields will fit all:

  family_name (Spring in 't Veld)
  name (Robert Spring in 't Veld)
Family name can be used for grouping and sorting. But you need to store the name as the user entered it. Don't try to split it in parts, just leave it as is.


You can't (simply) use family names to group Polish names (and I think Slavic in general), because spelling of names often (but not always) depends on gender, e.g wife of Mr. Kowalski is Mrs. Kowalska.


Yes: Slavic in general (my other example is Russian; wife of Makarov is Makarova).


Do you have any opinion on what the practices should be with names like yours? Should we somewhere between the view layer and the persistence layer convert to a canonical form? If so, can you provide any links for widely-preferred ways to do so?

About to find myself building some such systems and would prefer to do it right…


<rant>

Names are just Unicode strings, which you should not fold, spindle or mutilate. Why do you care which bit is the family name, or if this person has a family name, or even if there is such a thing as a family name in their family's culture? Why even try for a canonical form?

Even better, make name a _list_ of Unicode strings, so people who have multiple names (Chinese/Western, pre/post marriage, pre/post op) can give you the whole list to search on.

If asking for their credit-card details, ask "Name As Printed On This Credit Card" and store that as well so you can bill them. That's really a property of the card, not the person.

And then just add a "Informal Name" field so you're not stuck guessing which bit is the informal name, or writing "Dear Roderick Frederick Ronald Arnold William MacArthur McBan" which does sound kind of dorky ...

</rant>


One of the stupidest things we've done as a culture in the last generation, is bend our ideas about what is and isn't "well-formed", so that we don't offend computers. Fixing this is a big part of why I am motivated to work on schemaless databases. When your database doesn't ask you do predefine the data structures, it's one less opportunity for an ignorant programmer to blunt the human spirit.


The same thing applies to postal addresses. I look forward to the day when I can make my own shipping label when ordering goods online. This just needs to be a free text, multi line field.

The theme you touch on is covered at length in the book 'Computer Power and Human Reason'.


Shipping address has to be on the most shoehorned pieces of data in any database in the world. The only thing that seems to have put some structure on it in the US is the 911 implementations.

For most of high school years, our house was listed as "311 Behind the School" or some other variation. There were no actual street names (although I am told the electric company and only the electric company had a map with street names). When 911 service hit the area, we got actual addresses, but they really weren't useful for shipping things. The local UPS guy knew where everyone was and where they worked so he could do his deliveries. His replacement required some training time.


May I ask where this was? Crazy to me that there would be a place with UPS service but not E911 service.


Rural ND on a reservation. E911 was implemented not too long ago (5 - 10 years). UPS in ND was always good at getting you your package. The drivers knew everyone and where they worked (or knew who to ask). Packages sent to my home actually got dropped off at my work (in person, well except the furniture type stuff, then they would warn you they were headed to your house so you could put it in the house).

The Post Office didn't do a route on the reservation (come pick it up - all PO Boxes), and FedEx was the most pain in the rear, evil group ever (drive 2 hours to pick up you package because we stop delivering in late Dec). You can work up quite the mad driving two hours to pick up network cable in a ND winter.

Around 1996, I actually had to tell a vendor we couldn't do business with them anymore because they had to ship FedEx. FedEx did get better, but don't mix up FedEx and FedEx Ground else there will be a lecture.


We have Fedex/UPS/DHL and use directions in Nicaragua and surrounding countries - I live 50 meters south of a doctor's clinic.


Its amazing when you get in rural environments how directions like "left after the old Benson house" (never mind Benson has been dead for 40 years) or "first house on the left in the 52 housing" (52 was the year the houses were built.

// never mind the "Easter Egg housing" and how they got their name - cheap*$$ government and their "bright" paints


This just needs to be a free text, multi line field.

I tried that for a project, using the Google geocode API. Worked OK, though not flawless.

Users entered their address into a textarea. This got passed off to Google, and the app then looked at Google's response. In case of ambiguity the user would be prompted to clarify the address and try again.

The intent was to reduce crap in the application database, but at the same time we had to allow users to override whatever Google thought and provide a literal address.


Also "You Are Not a Gadget".


Well.. if the database doesn't enforce a schema explicitly, then the code will impose one implicitly. Maybe that makes some types of applications easier to develop. But the user doesn't get much say in the matter either way.


Whoa, easy there - many of 'us, as a culture' have been fighting this tooth and nail.


<tinfoil hat> Remember, Google is getting into the Social scene because it's a rich data set that they don't currently have access to, and Google likes data. Google wants your "real" name so they can tie all your information back together. The good news is, they'll probably ease the restrictions once they figure out how to normalize identities even with non-normalized names.</hat>

As a side note, I'm in a similar boat to Patrick with respect to my real name: at least 6 variations have been used on legal forms alone, due to length and special character restrictions, and excessive use of spaces. Fortunately it's only my middle name that gives problems so it can often be ignored, but I really like it so it's still annoying when I have to distort it.


If that was the reason, they could just have asked for your real name without requiring that it be your public name.


i don't see how having a real name is interesting to google. a handle is a much better unique identifier, which is what you actually want when you are attempting to collect data from a bunch of sources. i could tell google that my name is james smith, and they'd be happy but it would be absolutely useless (and false). i could tell google that my name is notatoad, and that would probably be enough to aggregate all the information there is about me online.


As much as I understand the privacy concerns etc., am I the only one who REALLY hates being on Facebook and seeing a fake name with all sorts of odd punctuation and random capitalization in a list that consists otherwise of clean "Firstname Lastname"s? Even seeing "Equality" as someone's middle name bothers me to an extent.

Maybe I have OCD, but I'm personally glad I don't have to see this on Google+ for now (but I wouldn't be surprised if they changed their policy on this within a week or two given how fast they acted in response to the "gender must be public" feedback).


I have a number of former students who do this, and generally they have come up with some clever way to modify their name so that a search on their IRL name will not turn up their FB page, but a glance at the FB name will immediately bring to mind the IRL name. It's a useful tactic for anyone job-hunting, honestly, because as inappropriate as I find it that employers take into account irrelevant outside-of-work behaviour, the fact is that a lot of employers do take into account irrelevant outside-of-work behaviour, and it's a perfectly valid privacy tactic to make oneself less net-stalkable.


I do this on FB by removing the vowels in my last name. It is clearly obvious to anyone who I am once they see my profile/listing, but searching my name will not bring it up.


I knew a guy in college that went by the name of "Rebel". "Rebel One" if you were an instructor. His real name, and I am not making this up, is "The Rebellious One" and yes, he changed his name to that. I'm not expecting to see him on Google+ anytime soon.


I totally agree. I can't stand people who put random crap in their Facebook names (and actually tend to unfriend them.)

Similarly, I religiously give everybody on my IM buddy list their real name. I would go crazy if everyday I only saw pseudonyms and had no idea who was who.


What do you do when two people have the same real name? Also, I personally only think remembering everyone's last names, in an area of society where you hardly ever use them, becomes infeasible once you scroll through your address book and find 15-20 people named "Michael" (this is the case in my phone): it is often much more valuable to name people to include their context.

(Which, I might say, is another "myth of names" to add to patio11's list: "the list of names someone goes by is objective and transitive"; I might think of that person as "that Steve from Hacker News", when other people call him "Steve Johanson", "Stevey", or even "stivox".)


I don't use Facebook, but many of my Google+ contacts are people that I've know for years on IRC. I'm going crazy because I can't see pseudonyms, and so I have no idea who's who.


Clearly you never used IRC. When I got started with computers, BBSes, etc., using your real name was considered a monumentally bad idea.


I still don't understand why the fetish for the names from government ID originated. Was it because of Facebook's policy? I'd like to see this trend reversed.


Does everyone have a first name and a last name? Or do you only see people on the list whose names can fit the mold?


Currently on my Facebook friends list:

  * A MTF transsexual who prefers going by a feminine spelling of her name
  * A Vietnamese girl who prefers to go by an adopted American name
  * A Jewish woman who prefers to spell her name in Hebrew, and whose legal name contains six names
For all the things Facebook gets wrong, the name input is really not one of them, compared to Google+.

People have known me as MAD for over a decade now, and although I sometimes wish I could just go by Corbin, it's something that I'm stuck with, better or worse. Nomenclature is hard, and any attempt to shoehorn names into a single culture's rules is going to fail.


I'd be fine with any of the examples you gave, actually. I think as long as they "kinda seem" like they could be real names and are properly capitalized, my craving for consistency is satisfied.


They are real names. Your idea of what constitutes genuine nomenclature is distorted and narrowed.


I wasn't saying they weren't real names (though we'd need to define what "real name" means to actually decide that). I was just saying that in order to fit into my personal model of "consistency" they only need to seem like they are real names.

Secondly, as I sort of hinted at in my first post, I acknowledge that it wouldn't really make much sense to design a policy around my personal visual preferences, which I realize are a bit culturally exclusive. Nonetheless, I can't help but be annoyed by some of the names I see on Facebook.


I guess google+ won't like Filipinos then: http://news.bbc.co.uk/1/hi/programmes/from_our_own_correspon...


I am disappointed by how software developers continuously reinvent the wheel badly. There have been comprehensive data models for human names available for years. For example see the HL7 V3 EntityName data type. http://www.hl7.org/v3ballot/html/infrastructure/datatypes_r2... Any Entity can have multiple names, each of which is tagged with zero or more use codes such as "official", "pseudonym", "maiden", "tribal", etc. Each EntityName has one or more ordered parts which can optionally be tagged with a type: prefix, suffix, family name, given name. This data model isn't fully complete since it's missing some naming concepts such as patronymics but at least it's a better start than the mess that Google came up with.


Though what would be better and easier to fill out: (a) multiple text boxes with one for each important kind of name, (b) multiple text boxes together with some sort of code selection for each text box, or (c) a single text box that just says Name.


The usual solution is to provide users with a choice. Default to showing a simplified field or set of fields that suit the common case for that locale. And provide an "advanced mode" with more fields and options to handle unusual cases.


Well you could do that, but only if you must collect such information for governmental purposes, otherwise I would stick with just the single text box and allow them to put in whatever name they wanted.

I would especially do that for any sort of web startup for both simplicity of the interface and ease of implementation.


It's not just for government purposes. If your hypothetical web startup has to integrate with any third-party systems (like anything in healthcare or financial services in the US) then you need the name broken into parts.


Well I would consider that to be for government (mandated) purposes, since the service you're contacting with has a government mandated schema for the name it must use, and hence you must also use.

Of course you can always have different fields for billing information.


In most cases those requirements for name format are not goverment mandated at all.


I still think that the most bothersome thing is that there are lots of people, especially in my generation, who grew up with the Internet and with identities tied to ancient AOL e-mail addresses, screen names, Usenet handles or MUD characters that are sometimes both more cherished and more important than their actual name.

What's odd is that my rule for social networks is that if you know my name online, I might want to connect with you. If you know my real name exclusively, I probably wish I didn't even know you.


I was kicked off Quora yesterday because they didn't like my 'Real Name' (which is actually my completely innofensive real name)


Everybody get over it, preventing anonominity and privacy (as each individual personally feels it), is a fools game. In the long run there will be no (as in absolute zero) acceptance of corporate need to determine appropriate levels of privacy. This is because people are not fools. Facebook is going down because they misunderstood how people feel about privacy, and a poor replacement (privacy wise) will meet the same fate. As the saying goes, you can fool some of the people some of the time yada-yada.... The shame is just in how much they leave on the table by trying to take to much, but nothing new, and on we go.


Requiring real names. Acquiring a facial recognition company. Profits entirely driven by advertising.

Well, hello Minority Report ads!


Let's call it for what it is: facebook's process/technology to determine real people is many many notches above google's.

This is as much tech failure as it is a policy failure.


As other posters have noted, many Chinese have the Chinese-lettered name eg "strong army", the pinyin version and if they move to a Western country they pick their own English name like "John" or something. In writing, Choi/Choy (and I think Chua and Chow) are the same. I will skip the philosophical discussion of what reality is to the viewer.

I don't understand the linking to a real name. Give everyone a private number, and then the user links whatever names they're known by to that number. When a searcher finds the name Skud and adds Skud, they will forever see Skud and whatever other names Skud has chosen to have visible to that circle. The link is the number, and the name is just a display. I wonder how they would deal with a woman who changed her last name when she was married, and reverted to her old name after the divorce. And she's also an author writing under a pen name (like Stephen King/Richard Bachman/John Swithen). It seems obvious to me but nobody's done it so there must be some unique flaw. Can any commenters enlighten me on why the unique private number idea is a bad one?


Why make it a unique number?

As Google+ has realized with circles, we have different associations we make in life.

Some of those are widely separate identities.

I see no reason why any identity I have on one site should be associated with one I have on another, if I deem that it not be.

For some purposes (voting, financial transactions, long-lived financial accounts such as SSI or a life insurance policy), you'd want to tie one identity to another. Beyond that, it's simply a control and surveillance front.

Even in the cases I've mentioned, weak authentication has long been the rule. Strong authentication in voting is often tied to poll taxes and other means of restricting the electorate. Public corporations (literally "anonymous societies" in French) are highly psuedonymous. Cash (and digital equivalents) are untraceable. And numbered bank accounts are the stuff of legend in both finance and noir literature.

Identity is a very, very deep, and frought, question. Curiously, G+ is turning into quite the discussion of it, from circles to gender to names to multiple identities.

One of the most classic instances of pseudonymity is among revolutionaries. It played a large role in the American Revolution, particularly among pamphleteers (the 18th century analog of bloggers): http://www.magic-city-news.com/Editor_s_Desk_34/A_Climate_of...

What of Mark Twain, Lewis Carroll, George Eliot, George Sand, Ellery Queen, Frank Dixon, and Carolyn Keene?

A particular usage is among revolutionaries: Lenin, Stalin, Golda Neir, Moshe Dayan, Subcomandante Marcos, Carlos the Jackal.

Or stage names: Madonna, Lady Gaga, Huey Louis, John Wayne, Marilyn Monroe, Bono, Cat Stevens, Yusuf Islam.

You're making an extraordinary proposal. Support your position.


Oh sorry, I'm wasn't intentionally arguing against being pseudonymous. I was thinking of ways to manage the mutability, multi-mapped and non-uniqueness of common names behind the scenes. If circles can be kept separate, you could have a single login containing different circles for your pseudonyms. As an aside, I hate how current FB and G+ policy turns us non-celebrities into second class citizens. Disgusting.

Thanks to nradov above for info on HL7. Surely these major international orgs should already have been aware of it?


"I am not a profile, I am a free man!"

http://www.youtube.com/watch?v=zalndXdxriI&NR=1#t=41s


What does this have to do with "web scale"? I don't see this phrase in either the article or the reference in the article.


The title is a reference to the fact that G+'s identity system fails to cope with names considered normal in their respective cultural contexts. In fact, it couldn't be more to do with "web scale" since it refers to how literally everyone in the world reveals their (sometimes multiple) identities through Google's world wide social network.


It might be a poorly defined term, but I believe "web scale" refers to the performance and availability, not features, of web applications. To say "They messed up feature X for population Y" doesn't speak to their ability to scale.

No matter how much effort Google put into supporting "real names", they would still disappoint somebody. Does Facebook do this any better? I believe they also require users to provide their "real name".


Despite how many people seem to claim that Facebook requires real names, I have numerous people in my list of friends that have never had a real name associated with them; these are not new accounts, nor are they inactive ones: these are real accounts my friends have been using for the last 8 years on the service. Facebook somehow "gets" this in a way that Google may be trying to "cargo cult" in some way, but if that's the case Google is missing the point and failing at it.


"Web scale" refers to dealing with the problems and issues that occur when you have hundreds of millions of global users. This includes performance and availability, but also security issues, social/cultural issues, and business/tax/legal issues. (Not that "web scale" is in Websters -- just google around and look at the other uses)


The naming problem is no greater proportionally 'at web scale' than it is at lower numbers of users. Used in this context, it's a cringeworthy bit of chin-stroking jargon.

At global scale would be a more sensible. 'At web scale' doesn't mean, 'With users from outside the US'.


re: proportions: If Google+ was only for Chinese plumbers, the percentage of "problem" names (and the kinds of problems those names posed to the G+ DB schema) would obviously be different -- hence the usefulness of the term "web scale" when talking about this sort of problem.

re: "web scale" not implying/talking about global users... Do you know of anyone operating a web scale business that doesn't have global users? Even sites that are supposedly for US customers only (netflix, etc -- caused by regional licensing), still have non-US customers using proxies to bypass their geo filter...

Web scale /is/ global scale. :p


Hmm, actually global scale is wrong too - I retract that :) The problem arises regardless of 'scale' (number of users).

To reiterate: the number of users (scale) is irrelevant to the issue of problem names.

I can see why someone sprinkled buzzwords on the HN title that didn't exist anywhere in the article, and weren't applicable though. Makes the issue sound more 'techie'.


I wrote that article, and consider the title I assigned to it here on Hacker News to be its official title; this is also the title that I used when I linked to it on my Facebook Page. Why, therefore, it matters at all that these words "didn't exist anywhere in the article" is beyond me.

As for "buzzwords", if you deal at all with Google employees, this becomes an irritating mantra: systems that have less than a hundred million users are often considered "toys" (and, to be clear, I can totally understand why this would be this way to these people).

And yes: I think that the point that "this actually happens to everyone everywhere" is an interesting point, and was the first comment on this post by patio11. I found that a very interesting insight, as many of my examples are more "those cases that come up when dealing with a global user community full of interesting edge cases", which to me /defines/ "web scale".


  Why, therefore, it matters at all that
  these words "didn't exist anywhere in
  the article" is beyond me.
What is the purpose of the quotation marks then? Irony?


Yes. Again, that is the official title of the work: imagine that it had been at the top of the linked to page, ironic quotation marks and all.


With the benefit of time and reflection, I now think that the term web scale is fine in this context and I was just being a curmudgeon. For what it's worth, which admittedly isn't much!


You should be so lucky.

My entire planet is forbidden.


Can somebody explain the problem with allowing people to choose their own names? They claim it is to foster "an environment." What kind of environment? Surely not an honest one, since using real names doesn't cause people to act honestly.

Forcing real names fosters an environment of rigidity and conformity. But I guess people like that about FB and Google+. At least MySpace allowed you to pick your colors.

The Internet could be it's own place, it's own unique environment, but instead people like FB or Google+ want some poor attempt at a one-to-one mapping of the real world.


NB: the nonsense phrase when operating "at web scale" doesn't appear anywhere on the linked page.


Given that I wrote both the article in question (on a medium, Google+, that has no title field) and the title of this post, I think it is fully in my right to assign the title the way I want to ;P.

For context, though, that phrase is there to point out that when you are dealing with a worldwide audience of tens to hundreds of millions of people, you cannot make silly assumptions (see patio11's set of "myths") regarding names.

As for using that /specific/ wording, that is how Google themselves often describes the state of operating with that large, and that diverse, of a userbase, making it anything but "nonsense" and in fact fully "apropos".

(Also... what is "NB" supposed to mean?)


Implication of "NB" in this context: don't blame the author of the piece for the cheesy buzzword in the HN posting.

I twig now that the author and the HN poster are one and the same internet celeb. So... sorry about that :)


For the most part, I think that Google's crackdown on pseudonyms and anonymity is actually a good thing. In most cases, anonymity online brings out the worst of us—just look at 4chan, littered with mobs of minions proliferating senseless hacking and child pornography[1]. You need but look at the news lately to see the damages of unchecked anonymity: "LulzSec this," "Anonymous that" and so on.

Enforcing real names is a good thing. It means people finally start to take responsibility for their actions, and there is accountability. People behave much better when the threat of embarrassment is in the balance.

I only see one legitimate reason for someone to be allowed a pseudonym: if they are more widely known by that name than their real name. This applies to authors, artists as well as web community members. The solution is easy: allow a nickname field in addition to your real name: [First] "[Nickname]" [Last]. Some already do it. Day9, for example, goes by Sean "day9" Plott online.

Lastly, I realize the hypocrisy of posting to Hacker News without my name visibly attached, so for the record, I am Kenneth Ballenegger from kswizz.com.

[1]: I have browsed thru /b/ many times, and the behavior of people there truly is the worst I've seen ever. I was in the middle of the SF Giants riots last year, and the people setting fire to cars and breaking windows seemed more civil by comparaison.


I guess I haven't seen much correlation between real names and quality; more seems to be between the forum type and quality. I'd be interested in a study of that, though.

Two random examples: HN is mostly pseudonyms and has pretty good quality discussion; my local newspaper now uses Facebook Connect to have users post under their real names, and its readers dutifully post a bunch of ignorant trash under their real names.

I do like having semi-stable identities, whether pseudonymous or real. It makes more of a sense of community if you actually recognize people and ascribe viewpoints/personality/etc. to them, rather than every post in effect being written by RandomUser92736 with no continuity between discussions, or any ability to form within-community reputation.


I think you're on to something here.

There are good communities and bad communities.

Norms play a huge role. Expectation plays a role, Participants' culture plays a huge role. Some communities work well based on in-real-life associations. Some work based on technical means (strong moderation tools). Some have weak moderation tools but an aggressive enforcement community and back-channel discussions for dealing with egregious abuse, though from my experience those back channels are used far more rarely than might generally be expected, and may even take on an inside-joke status.

In the same sense, real-world communities may or may not be well-behaved, and often have similar foundations. The better-behaved ones tend to be well-established (but often hard for outsiders to break into), highly policed, based on cultural norms (common religion, ethnicity, purpose, social class, interests). It's where multiple structures break down that problems become most severe.

Even short-lived communities can be very well behaved. There are few communities more transitory than college campuses, yet they have very few management issues.

The biggest hazard in either online or physical cultures is creation of a mob (root word: mobility), in which an antisocial mindset spreads through a population. What are needed aren't strong authentications ("papers, please"), but circuit breakers and dampers to dissipate these energies.

If it is in fact civility and community which are valued.


4chan is an example of an "anonymous" system, not a "pseudonymous" one: the difference is whether you have stable identities over time, and therefore can build up reputation (positive or negative) for a particular "pseudo-name".

As long as you have mechanisms, whether social or political, to "default distrust" new users that don't have much "cred" yet (mechanisms like the "green" user rendering that Hacker News uses for this purpose), you are not going to lose the benefits of "anonymous" vs "real names".

In fact, one might even argue that claiming otherwise misses the point of names: even when I am going by just the name "saurik" (seldom, but it happens: like here on this site), I care deeply about what I say as it affects my identity.

As an example near and dear to many people on this site, no one had known who "why the lucky stiff" was (maybe they do now? not keeping up with that much); but, if he had started being an idiot on forums and started "proliferating senseless hacking and child pornography", you can be certain that that would have been quite bad for him.

(In fact, if your real name is not known by anyone, and everyone knows the entity that actually has all of the reputation capital you draw on on a daily basis, then arguably there is no risk in doing something stupid with your real name, and total risk in doing so with your pseudonym.)


I agree with you on two points. Firstly, you are right to make the distinction between anonymous and pseudonymous, they're two different things and you can build a reputation system and healthy community on a pseudonymous system (see HN). However, I think they both exhibit the same underlying problem that comes with lack of accountability. HN has been lucky, because it's for the most part an elite community full of smart people. Digg or reddit have not been so lucky, and the quality of comments there is quite low. HN is the exception, not the rule.

Secondly, I also agree that there's definitely some edge cases where it make sense for people to be anonymous, or pseudonymous. If you need to shield yourself from your employer for legal reasons, I think it makes sense for you to do so. What I'm arguing is that in most cases, though, it's good for the overall community to enforce true names.


I use a pseudonym to shield myself not so much from my current but my next employer. Or landlord. Or girlfriend's family. HN gets what's really weighing on my mind, while Facebook gets heavily censored, cheerfully inoffensive pablum when I bother to participate at all. As I see it,

  true names : pseudonyms :: villages : cities
and cities are the wellsprings of progress because people can reinvent themselves and unpopular but good ideas can find support.


There's more to pseudonyms than legal reasons. What I write on the internet stays there, and I expect people to google my real name, but I do not wish to have all of my lives linked together.

For example, if I'd have to use my real name, I would never speculate. With a pseudonym, I feel comfortable making mistakes and tiny contributions. Similarly, if I were in a horde guild, I would not want alliance players to dismiss my civil politics opinions.


I have browsed thru /b/ many times, and the behavior of people there truly is the worst I've seen ever.

This has little to do with anonymity. Millions of people act equally stupidly on Facebook or Youtube as well, where there isn't such anonymity. The quality of posts is determined by the quality of the community and its members, not whether or not they have names.

/b/ is hardly representative of anonymity on the internet, and if there wasn't any anonymity, /b/ would be just as bad as it is today. Even on 4chan, the vast majority of the boards have a vastly higher quality of posting than /b/.


So should in your view Wikipedia and GitHub enforce real usernames?

You make it sound like there is no responsibility and accountability in pseudo-anonymous cooperation.

How about real names on license plates? X-Box live accounts?

Enforcing real names is _not_ always a good thing. They already have my IP and the police can demand my ISP for my personal data; not just a stalker, impersonator, or any fraud.

Enforcing real names on 4chan will solve nothing, and likely make matters worse (in your view): It would create a more vibrant Freenet or Darknet where people are a lot more anonymous and 4chan seems civil by comparison.


People love to bring up 4chan as an example to this discussion every time, and yet there is no one on 4chan who has not freely chosen to be there.

All this "know who you are talking to in order to have a friendly community" stuff is nonsense. We know why Google and Facebook and the Obama administration want this. Control and tracking. End of story.


It's not about what 4chan does to its members—I couldn't give less of a fuck. It's about what they do to the rest of society. Things like Anonymous and LulzSec, which end up on the NYTimes frontpage, are is a direct result of places like 4chan.

Additionally, as a member of the Google+ community, I'm glad to see Google cares for its quality.


Things like Anonymous and LulzSec, which end up on the NYTimes frontpage, are is a direct result of places like 4chan.

I must disagree with this statement. If you believe that anonymous boards like 4chan provide a gathering place for the already antisocial, getting rid of 4chan doesn't get rid of the antisocial. They'll find some other way to gather and/or express their frustrations with society. I'd argue that it's possible that sites like 4chan provide an outlet for some people, preventing them from expressing their frustration "in real life."

For the record, I consider the real-name-only position you've advocated on this article to be a bit extreme (I believe I also replied to another of your comments). Feel free to desire a place, web site, etc. that satisfies every one of your whims and expectations, but don't do so in a way that forces others to conform. Maybe they see value to themselves and to society in the things you personally dislike.


Anonymity is not related to lack of forum moderation or escape from consequences. It just means a username is not recorded or displayed. BTW, posting underage obscenity, inflammatory statements, neo nazism etc gets the poster ip banned. [1] /b/ is an isolated case which is often quoted. It is only one of 40-something forums on 4chan. Please, go back and browse /adv/ instead. Same anonymous people, supporting and helping each other with advice. Anonymity is not the issue, it is the community which people create through their actions and expectations of finding other like-minded individuals in the same virtual space.


The difference in discussion quality is this:

With identity, because users can form a social hierarchy and praise or shame each other, they live in a box that others have defined them to be in. You may be in a pleasant box, but you are not seeing "the truth." Everyone manipulates their image and how they present themselves to others. This is fine within small groups as it can make the group more stable. However, it does not work at scale; this is the subject of Assange's theory of "conspiracy as governance." Blow the whistle on the conspiracy and you will be a threat, and shut down by any means necessary...

Without identity, you have to constantly prove your worth. What Anon tells you is incoherent, and usually overly emotional, but also usually honest...trolling aside... And in this way, truth can gradually be exposed.

Since Anon's breakthrough into "the real world," governments face a situation where the global strong-identity hierarchy is being disrupted. They can no longer spin a manipulated version of facts to the public and expect it to last for very long.

We need some of both. Strong identity within small groups; weak identity when dealing with society at large.


So is there no good reason one may wish to remain anonymous or pseudonymous? Because I can think of many.

-- Chris Sauro, masterzora.net


LulzSec are vandals and dicks. But Anonymous has done some good things as well...

> [1]: I have browsed thru /b/ many times, and the behavior of people there truly is the worst I've seen ever. I was in the middle of the SF Giants riots last year, and the people setting fire to cars and breaking windows seemed more civil by comparaison.

Seriously? Yeah, /b/ is a vile cesspool. But there is a real and significant difference between riots that directly affect people's safety and destroy physical artifacts and capital as opposed to information leaks and DDOS attacks.


There is a difference between anonymous and not claiming responsibility for what one says, and being anonymous and claiming responsibility for what one says. 4chan is an example of the former, while the Federalist Papers are an example of the later.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: