Comments

AnonEMoose • January 6, 2017 10:22 AM

Facebook > 4chan/b/ by far, in terms of reputability and behavior of its members. But that doesn’t mean there is unreliable junk on Facebook.

Not my real name • January 6, 2017 10:44 AM

This fits with my observations over the years.
On facebook, I see people I know posting hate-filled (usually racist) things under their real name that they would never say in person.
In email exchanges, I see people write trollish things they would never say. The online disinhibition effect is independent of anonymity.
Several months ago, the Canadian Broadcasting Corporation got fed up with the amount of racist abuse being posted in their website’s comments section, and introduced a “real names” policy. The result: The comment threads went from being somewhat toxic to being completely unbearable. Voices of reason or moderation became much scarcer, and bigoted trolling ended up dominating most of the comment threads.
A weakness of the “real name” policy is that anyone posting reasonable or moderate opinions online faces a real danger of reprisals in meatspace.

Evan • January 6, 2017 10:48 AM

@AnonEMoose

Not really. You assume Facebook is better behaved because a) most of Facebook’s content is privacy protected, and b) you can attribute bad behavior you do see to an individual person like “Bert Bertelsmann” rather than a faceless group like “Anonymous” or “4chan /b/tards”. That kind of cognitive bias is exactly what the article is about.

Joe Patterson • January 6, 2017 11:08 AM

Gotta laugh at using real names. I wouldn’t be surprised a bit if there are thousands of people with my name (based solely on the number of them that think they have my e-mail address. Thank you Delta Airlines for informing me in great detail about every step of Jason Patterson’s flights last week). I have privacy protection through name-chaff. And, similarly, plausible deniability for anything attributed to me.

Daniel • January 6, 2017 11:27 AM

One thing that I have observed is that if someone uses a handle that could be real name then readers assume it must be the poster’s real name. For example, “Spaceman Spiff” is unlikely to be someones real name and so no one thinks it is. On the other hand “Daniel” is a real name and so people tend to assume that it must be the poster’s real name, when it is in fact a pseudonym.

It’s an example of how people confuse the concepts on anonymity and camouflage.

Bong-Smoking Primitive Monkey-Brained Spook • January 6, 2017 11:43 AM

And this is not my real name, just so you know. It’s used for vulgar and politically incorrect content. I could post under my real name but I don’t want the headache.

Can you imagine Gupta or Mordechay reading this comment, and I end up interviewing with any of them?

Gupta: I have good news and bad news, Bongo! You meet our technical requirements expectations. But that crack… I don’t appreciate. Take your resume (or business) elsewhere. Best of luck! By the way, I’m very well connected, and the word got around, hehe!

Kiyoshi Aman • January 6, 2017 11:47 AM

As a rebuttal to the article’s assertion that a bit of anonymity encourages better behaviour: YouTube comments. Further rebuttal: comments on news articles featuring particularly heinous activity.

Wendy M. Grossman • January 6, 2017 12:32 PM

It really baffles me how this myth persists. All the way back in 1988, Sara Kiesler wrote an entire book about the bad behavior on internal office emailing lists, where everyone was identified and at work. In the early 1990s, “flaming” was known to be an endemic part of online communication. C. 2005, Danielle Citron, author of Hate Crimes in Cyberspace and of a recent Lawfare posting that suggests crowdsourcing troll identification, led a panel at Computers, Freedom, and Privacy on online bullying. More than one speaker talked about professional mailing lists they were on where everyone was fully identified and told stories about behavior so bad that in a couple of cases they called the person’s employer to complain.

wg

Matt • January 6, 2017 3:48 PM

the irony I see in this is that if facebook didn’t try to imply that they care about using real names/make it so much of a “MEATSPACE PROFILE” (they don’t, people use pseudonyms/create fake profiles lately), I wonder if facebook would be as popular or as filled with uselessness as it tends to be.

albert • January 6, 2017 4:00 PM

@Spaceman Spiff,

I thought ‘Spaceman Spiff’ was an anagram for ‘Pace S. Spiffman’, I girl from my HS class. I am so disappointed. She was cute.

. .. . .. — ….

. • January 6, 2017 5:05 PM

People also respond differently to a disembodied voice conveying impersonal information than they do to a name or nym they’ve gotten to know. Responses are less biased by the ‘relationship’ so cognitive dissonance comes out more readily. That unvarnished response is much more useful than pseudo-social politesse. It tells a lot about the source of resistance to unfamiliar ideas; also it immediately signals the degree of groupthink or epistemic closure in the forum.

As in meatspace, demands for identification are just another authoritarian control neurosis.

@albert • January 6, 2017 7:50 PM

:s/@/re: /
[(Probabl|Mayb)e/?]?+

What did I mean? Oh wait, nevermind. just lost my spliff again

Nick P • January 6, 2017 11:08 PM

@ Bong-Smoking

The door size was funny. The Rowhammer joke at the end I thought I got but checked on it just in case I learn something new. Then I see this:

“Consequently, read operations are of a destructive nature because the design of DRAM requires memory cells to be rewritten after their values have been read by transferring the cell charges into the row buffer. ”

Perfect example of abstraction gap like the logical vs physical version of diodes. People are told you read from or write to RAM. When you read, you get the value from RAM without affecting it. Reads are free and can’t hurt. That’s why many security schemes, including the software versions of mine in the past, allow reads but not write. In physical reality, a read is a write for DRAM. The abstraction is a lie. Damaging attacks resulted.

I wonder if there are variants of it where reads are truly reads. Writes doing just what they need to is probably just a QA issue. The whole attack is probably a QA issue but I’m talking a design strategy with better potential at matching abstractions to reality. 😉

Clive Robinson • January 7, 2017 5:50 AM

@ Nick P,

The use of CAS/RAS with DRAM goes back atleast as far as the 1970’s from my memory.

The story behind the DRAM refresh is that not only does it have to be done very regularly an entire row has to be done at the same time.

Due to the nature of the circuit any read is destructive of the storage cell contents. So the entire row is read into amplifiers with latches known as “sense amps on every column line act as temporary storage. The output of the latches is read by both the IO pad amplifiers and then the row write amplifier to restor the correct levels on the storage cells (the actual refresh).

In theory you could only do a refresh on a write, but after a second or so’s thought you will see that that is not practical as the bulk of main memory is effectively “read only” in it’s main use, you would have to get the CPU to perform a read followed by a write on every memory location that would not be practical.

Originaly you had to build your own refresh circuit using counters and logic that interfaced with the CPU address lines and Read & Write lines. This fairly quickly got incorporated inside the DRAM chips and over the years became more complicated in it’s ability to perform the hidden read cycles.

The thing is there realy is no need to use DRAM any longer SRAM is faster and the silicon real estate differential is not as large as it’s made out to be by the pricing differential. Quite a few people have accused the DRAM manufactures of running a price fixing cartel because of the overly high prices of DRAM. If there is a cartel or not I don’t know, however the pricing is certainly skewed compared to other memory technology and CPU technology. Why this has not been investigated I have no idea… But there are way to few DRAM manufacturers so the market is badly skewed because of it. Likewise the cost of entry is astronomical, and you would need a vast amount of external investment. Thus there is a very real danger if you tried entering the market, the encumbrents would wait untill an oportune time to drop the pricing to bankrupt any new entrant before their product became available. Thus investors are not likely to come forward to provide the financing…

Nick P • January 7, 2017 9:15 AM

@ Clive Robinson

I hear mixed things on DRAM vs SRAM cost. Seeing your statement, I decided to just look at the circuit or physical layout of each to compare. After developers & tools, the main per-unit cost of hardware is the physical area it takes up. The SRAM circuit I found takes up significantly more memory with Cypress’s description sounding right: 1 transistor and 1 capacitor for DRAM vs 6 transistors for SRAM. That would be three times the area which leads to an upper bound on three times the cost per unit.

At that point, determining whether replacing DRAM with SRAM was feasible would require getting together all the control, monitoring, I/O, etc additions onto the RAM stick DRAM already has plus squeezing a pile of SRAM into a chip of similar size. The resulting area tells us the cost. That multiplied by however many chips is on a stick plus the cost of the stick is the final price.

Clive Robinson • January 7, 2017 12:19 PM

@ Nick P,

Cypress’s description sounding right: 1 transistor and 1 capacitor for DRAM vs 6 transistors for SRAM.

Careful how you count things transistors in SRAM are getting smaller all the time, but that DRAM capacitor has to be a certain minimum value depending on the refresh time, thus it’s not shrinking in size. Further there are other issues that stop that one transistor in DRAM shrinking as fast as other transistors…

But as I said there is quite a bit more involved with DRAM than just the addressing of “bit cells” and that takes up space.

You actually need to get very down and dirty to make a proper evaluation of required area.

MikeA • January 7, 2017 5:36 PM

SRAM does not have to be 6 transistors. I recall using a 4 transistor cell (IBM VLSI process) over a decade ago. Also, as Clive points out, those capacitors are huge (in relative terms), and very clever stunts have to be done to work around that.

BTW: I used to have a couple memory boards from a flight simulator (back when that involved hardware, not just code running on MSDOS), that were based on unusual DRAM. It refreshed the entire array on any write. I found a datasheet and app-note that explained, but have lost it since, so any pointers gratefully accepted if I jogged someone’s memory.

Now: Refresh. Harry Husky once gave a talk about the design of SWAC, including an interesting issue. They used interleaved banks of DRAM (well, William/Kilburn storage tubes, but you can think of them as hot, glass, analog addressed DRAM) so they could “hide” the refresh of a bank during an access to the other bank. Interleaved on the LSB of the address so the normal (three out of four) alternation of the PC LSB, they would get enough free time. Then somebody coded a JMP SELF…

MikeA • January 7, 2017 5:44 PM

SRAM does not have to be 6 transistors. I recall using a 4 transistor cell (IBM VLSI process) over a decade ago. Also, as Clive points out, those capacitors are huge (in relative terms), and very clever stunt have to be done to work around that.

BTW: I used to have a couple memory boards from a flight simulator (back when that involved hardware, not just code running on MSDOS), that were based on unusual DRAM. It refreshed the entire array on any write. I found a datasheet and app-note that explained, but have lost it since, so any pointers gratefully accepted if I jogged someone’s memory.

Now: Refresh. Harry Husky once gave a talk about the design of SWAC, including an interesting issue. They used interleaved banks of DRAM (well, William/Kilburn storage tubes, but you can think of them as hot, glass, analog addressed DRAM) so they could “hide” the refresh of a bank during the access to the other bank. Interleaved on the LSB of the address so the normal (three out of four) alternation of the PC LSB, they would get enough free time. Then somebody coded a JMP SELF…

My Info • January 8, 2017 9:42 AM

It is game over for “average folks,” i.e. anyone who is on the Internet with a mainstream web browser and a mainstream operating system on COTS hardware, i.e. basically anyone with the capability of posting on this or a similar forum regardless of their level of expertise.

The “bad guys” already know what your real name is on the Internet.
The “good guys” just don’t care.

Bad guys (from Paris with love) • January 8, 2017 3:34 PM

@My Info

Hello there. We have spent three hours chasing you to tell that we do not care either. Do feel free to watch your wacky sick porn online as you are used to.

Office emails commentary spot on. Gives a fair amount of ballache when ‘leaked’ properly.

CarpetCat • January 8, 2017 6:37 PM

Where are some of the other notable fake names on this blog? Here we are, WAY off topic discussing ram voltage and refresh rates. Abby PressHereForCondescension, moderator, clive robinson*, ain’t nobody using their real name!

On reflection, using real names does prevent multiple handles. You would know who the mod is, and he/she couldn’t be a poster as well. I wonder if Bruce has an alt account…

  • -stay warm and try not to get sick this winter, (online) friend.

Barney Laurance • January 9, 2017 4:37 AM

I don’t think using real names does prevent multiple accounts per person. There are many millions of people walking carrying identical names on the documents in their wallets, e.g. John Smith or Muhammad Ahmad.

And it’s also not true in general that an individual only has one ‘real’ name. Plenty of people use multiple names in speech and in filling in official forms.

Clive Robinson • January 9, 2017 5:29 AM

@ Barney Laurance, CarpetCat,

There are many millions of people walking [around] carrying identical names

I do know that there are atleast five people in the UK Tech industries have the same name as me, and I’ve met some of them in person.

Worse perhaps from a security aspect is that I know people who actually look like me, sufficient for even those who know me to make mistakes.

Some years ago I went on holiday with a group of friends and we stayed at one of those “surrender your passport” hotels. On checking out, the girl behind the desk was not overly fast and the transport to the airport was waiting on us. She handed us our passports and we stuck them in our pockets and got to the airport in just sufficient time. Any way to cut a long story short, we got home safely and a day or so later I got around to putting my documents away, to discover I had the wrong passport, phoned my friend and he was likewise as supprised to discover he had mine. We met up and worked out we had both travelled all the way back from the hotel on each others passports… Which should tell you something about the inability of even people who’s job it is to check faces and photos to make even an approximate guess at a match…

@ CarpetCat,

stay warm and try not to get sick this winter, (online) friend

Thank you for the wishes, but sadly I’ve already failed in that endevor, my son had a bug, which he gave to his mum and me. Of the three of us I appear so far to have got off the lightest, in that all I feel is drained, with a few aches and chills. Both of them have been bed bound with outrageous “man flu” symptoms including soaring tempratures. I guess it’s there own fault they both had flu jabs, where as I as usual avoided it, or maybe with all my other problems I’ve not noticed 😉

Curious • January 9, 2017 6:10 AM

I just want to point out that my “behavior” is to not post at all on a forums threatening with banning the user if not using their real name, like norway’s national online newspaper Aftenposten.

I am also somewhat conscious about the possibly of being profiled by state actors, so I like to not just spill out details about my life for no reason, state actors who would probably claim in the media that they aren’t interested in what people say or do, but probably insisting on processing or storing user data and general content regardless, so they would end up lying to your face this way.

All media would probably do good to review their whole commenting feature. I don’t want to see blatant censorship, nor do I want to see a commenting field with maybe thousands of comments with little or no option for sorting through the comments in a meaningful way.

I shouldn’t have to use my real name to tell about my negative life experiences, because I’d argue that it makes the already bad commenting features in online newspapers obscene (a problem with little or no moderation because of lots of blatant censorship by deleting entire posts). As if pretending that a veneer of authenticity is key, but with dignity of a human being, and an extension this kind of goodwill similar to friendliness towards an individual being of no importance on behalf of the forum owners.

Then there is the issue of having a newspaper decided what things you are actually allowed to comment on. Presumably, commenting on some news about a war would not be allowed, not for commenting on sensitive news articles that might as well read as propaganda, by sort of demonizing, mocking or belittling people and other countries.

The way I see it, online newspapers that feature a comment field, is an obvious threat to free speech practicies, as if imposing this kind of censorship with demanding real names, with a reservation of deleting comments for whatever reason, and presumably cherry picking what articles a reader can comment on.

I bet when the monarch eventually dies in norway, the country will end up being some loony right wing country. These days the police patrol with a gun on their hips and you may find yourself standing in the queue line with an armed policeman in front of you when you buy food. And, the whole lack of housing issue, seem to consist of ever lasting price increases, and apparently speculators are allowed to buy up apartments, with prices rising maybe at much as 25% each year or so.

Not my real name • January 9, 2017 11:16 AM

Is obviously not my real name.

I never subscribed to the idea that real names changed behavior. Furthermore, unless you’re going to check my identity papers (comrade) that how do any of these sites know it is my real name in the first instance?

It is a futile exercise.

vas pup • January 9, 2017 12:41 PM

Tag: psychology of security

New AI development could be utilized for ‘instant’ security check as well(I guess it was utilized kind of in Israel for airport security already):
http://www.bbc.com/capital/story/20170108-speech-analysis-could-now-land-you-a-promotion

Speech tests are being used by some companies not just for recruiting, but also to assess and train communications skills, judge suitability for promotion, and to gauge employee stress levels. In all cases, nobody — at least, no human — is listening to what you say. But is it truly objective? And what are the risks?
It works like this. Your 15-minute voice recording is analyzed digitally — tone of voice, choice of words, sentence structure — to determine personality traits such as openness to change, enthusiasm, empathy. In a fraction of a second, a software program sums up your character. Charts and diagrams reveal how friendly, status-driven or well organized you are — compared to the recruiter’s ideal profile.
“There is no person in the world who would be able to analyze so many aspects of personality, skills and speech in just 15 minutes,” says Mario Reis, co-founder of Precire Technologies in Aachen, Germany.
The computer decodes your voice file, breaking it down into 500,000 aspects of speech. Your file is then deleted. When used for health applications, it is designed to be anonymous, without an identifying pin number, says Reis.

Bernard Marx • January 11, 2017 1:56 PM

I, for one, pick my real names from dystopian literature. It is like a sport.

Greetings from the Directorate of Hatcheries and Conditioning.

Dennis • March 16, 2017 5:45 AM

The fallacy lies in our intuitive assumption that without anonymity comes immediate accountability and hence an increased social responsibility to be “good” or “better”.

But as one of the mentioned researched states: “on average, people are actually more sensitive to group norms when they are less identifiable to others”.

There’s a valued intrinsic benefit in being a part of something bigger than yourself and contributing to that.

Reddit is a great example of that, even though there are a lot of different communities (subs) members are expected to follow the rules of reddit and of each community. Following the rules is rewarded by the community, and breaking the rules is punishable:

  • directly by other members (by downvotes)
  • by moderators (by temporary and permanent bans)
  • and by reddit admins (by banning and shadow-banning of accounts)

James • March 20, 2017 1:55 PM

You hit the nail on the head spot-on Dennis, Reddit is a great example but so is Yoat (reddit alternative) the whole reason it was created was because of the censorship on reddit. People want anonymity and they’re willing to go to extreme lenghts to get it.

Names though, I think we should be able to decide weather or not we use our legal government issued names (the ones our parents gave us) its just online once it’s there, it’s there forever which is why we seek anonymity is because we’d like to voice our thoughts without judgement, everyone is entitled to privacy.

I don’t see no difference between online or offline, if you don’t want to use your real name, then don’t but don’t enable people to force us to. Thanks for sharing Bruce!

Leave a comment

Login

Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via https://michelf.ca/projects/php-markdown/extra/

Sidebar photo of Bruce Schneier by Joe MacInnis.