the ads for phones which 'integrate into your social network' completely baffle me.
It seems like Twitter happened to be the latest online craze at the point when a bunch of politicians (especially in the Republican Party) decided they needed to embrace the internet, so they went with it and others followed.
I somehow doubt the world leaders themselves are actually Tweeting.
3: Some of Medvedev's early ones (he's a very recent convert) seemed like they were plausibly him.
3:Most likely it's some underling with the phrase "New Media" in their title.
One LOL is a tragedy; one million is a statistic.
||
We're getting walloped by a storm right now. I heard from a friend that this giant local whale tail is actually split in two, which is pretty nuts.
|>
We've already talked about the jerk in Utah tweeting about firing-squad "justice" (I think) and that pretty much sums it up for me. Though I confess that I'm on twitter and also dedicating these exclamation points to Stanley!! With so much love!!!!
Did you click on the same Verified World Leaders link that I did this morning?
11: No, what was that? I saw Medvedev news on this Ambinder thing linked to by TLL on the BLL thread.
Dmitry's tweets are kind of endearing; I wonder if it's because of the translation. Someone who can read Russian should read the Russian versions and report back. His latest:
Haven't had a burger in a while. Lunch with Obama at Ray's Hell Burger: http://eng.kremlin.ru/photo/177
#breaking: Character count constraint leads to exclamation point bloat!!! #paradoxical
Character count constraint leads to exclamation point bloat!!! #paradoxical
Maybe some people are under the impression that every "tweet" must be exactly 140 characters, and are looking to fill some space.
(Scare quotes around "tweet" because I, myself, am not a "twitterer".)
12 - on my twitter homepage, on the right hand sidebar, below the number of following/followers and above the @asilon, there's a rotating Link Of The Moment. This morning it was Verified World Leaders, and I clicked on it and laughed at various leaders saying, "Wooooo, let's have a big hello for President Medvedev!!!! Yay!!!!!"
Since I'm confessing, I'll say that I had to train myself out of wiming for 140 exactly because I try not to indulge my satisfying OCD tendencies. And also I'm not reading or writing particularly well now because my eyes are still dilated, but it's boring to do other things even when I can't see.
And Stanley, as my penance I'll offer that the way to get at me is to touch my neck or even talk a lot about necks or adam's apples. Ewwwwwww. I also get queasy about coprophagy and the human centiewwwwwww, but I think thwt's a fairly normal response. And I hope the storm has passed! Windstorms can cause amazing damage.
No one expects the human centiewwwwwww!
The forecast for today here had a chance of severe thunderstorms, and it was looking stormy for a while, but it looks like it ended up passing us by. Presumably the same storm.
19.last: Triggered by the passage of the same front, anyway.
||
Goddamn MS Word. I'm doing a long answer, which means numbered paragraphs, and there's something going on where Word starts numbering again from the middle sometimes when I edit a paragraph. I can fix it pretty easily, so I'm not worried about the final product, but it makes editing really confusing when all the paragraph numbers are wrong. Grr.
|>
I'm not sure Medvedev counts as a leader. He might be producing his own tweets because Putin won't give him a New Media Czar.
||
Man, I just spent like a minute squinting at the guy in photo #14 here, convinced that it was Joe the Plumber, before looking up Mitch Landrieu's photo.
|>
Composing text in MS Word considered harmful.
18: The human centiewwwwwww is playing just up the block. I shall resist. Really, I will. Besides, the central idea, to maximize the use of food, doesn't hold up.
But it's 100% medically accurate.
You could just buy one of these instead.
I guess twitter isn't going to go away. Or improve. I kind of feel sorry for the LOC.
26: I had somehow avoided a full mental picture of how it worked until now. Thanks.
I suppose I should just acknowledge that I'm going to keep using it too mostly to follow feeds, and finally install one of those special twitter application thingies. But I really don't want to have to run something other than a browser to use it. Any recommendations?
26-28: I pity y'all. I have perhaps 10 or 12 years left with that memory stuck in my head. You'll have many decades with it.
I assume this (Human Centipede: The Game!) has been linked here already?
OT: May I ask a favor? A few weeks back, Armsmasher, in the midst of explaining how he hates all of the world's children, suggested a particular microscope that might be suitable for my son's birthday. I thought I wrote down his suggestion but can't find it. And now I can't find the thread either. If someone could direct me to Armsmasher's comment, I'd turn my frown upside down.
I hope you don't belong to a discipline that requires sifting through a lot of primary sources or archive work, ari.
35: it used to be quite a bit cheaper. I suspect they are responding to evident demand.
27: I kind of feel sorry for the LOC.
What's that about, really? I mean, come on, if you're going to archive online data, your first thought is tweets? Really? I honestly don't get it.
36: No, firefighters don't really have to spend too much time with documents, fortunately. Thanks, neb.
38: Twitter does have the advantage of centralization. Most other online technologies have a variety of providers, which would make any comprehensive effort to archive them enormously complicated at best. At least with Twitter you're just dealing with one company which has all the data.
The same applies to Facebook, of course, but there you run into privacy issues. Another advantage of Twitter is that the data is already (mostly) public.
So the thinking is, or was: this is easiest, even if 99% of it is junk, so let's do that?
Advantage over what? It's not like there was some comparison decision making going on, as far as I know.
Top Twitter "emoticons" over 3 + years from here (an interesting site for datamongers):
13458831 :)
3990560 :d
3182129 :(
2935301 ;)
2082486 :-)
1461383 =)
1439234 :p
1013758 ;-)
979947 (:
669086 xd
I don't think I even know what :d means.
47: Smoking and eating a donut at the same time?
I don't think I even know what :d means.
It depends on whether you're in command or insert mode.
I don't actually think twitter is invaluable* not valuable, just that I think the whole public collection issue was handled too quickly. As far as I know, very few details had been worked out when they made the announcement, which is a lot of commitment for not really knowing what you're getting into. I haven't paid attention for months, so maybe it makes more sense now.
*Wait, "invaluable" means valuable? What a country.
I should say that I don't really know the details either.
On a very general level, though: On the one hand, the LOC is in the business of archiving lived creative output. Okay, and that has been very valuable. On the other hand, it doesn't archive every conversation overheard by people walking down the street. There's the puzzle about archiving twitter.
vi sucks!
Is there a capitalization issue here? :D seems more likely.
There's a decent argument for taking all of the twitter archive, if you're going to take it at all. There's so much in it that you can't really selectively choose which to keep and which not (unless you're prepared to write algorithms for that and then justify them, but we're not there yet). And unlike other ephemera-type things (like postcards, for instance) where stuff is sent on a third-party-designed medium, the medium-makers get all the messages too. The analogy obviously doesn't go very far; twitter isn't exactly ephemera (which is how postcards are often saved for their postcardness) or correspondence (which is how postcards are often saved for the communicativeness).
I take this to be what teo was getting at above; I was just objecting that the LOC doesn't seem to have been doing any comparison shopping, so to speak. Twitter made the initial approach and then things seem to have happened very quickly.
prob small d cuz ppl on phones dont waste chars
On the other hand, it doesn't archive every conversation overheard by people walking down the street.
Or rather, waste the time it takes to create a capital letter. Tweeters are busy people - like world leaders (except Medvedev).
53: Okay, I'll buy it. As a first foray into archiving the 'net. You'd have to take it all, I see that.
Before it's even happened, though (I think the twitter acquisition is still in the statement of intention stage?), I must say that lesson one seems to me to be that just as there are criteria, sifting mechanisms, in place for physically existing records (books, postcards, papers, audio recordings), so there simply have to be some developed for electronic records.
I must surely be stating the obvious. What a hell of a project that would be.
Why wouldn't the LOC take everything they can get their hands on? Storage is cheap. Search technology is ever better.
Storage is actually not very cheap for this sort of thing. The LOC, in theory, keeps things forever, which means lots of investments in backing up, long term commitment to migrating data or whatever the hell they have to do to keep it usable for, well, ever. It'll get cheaper, though. And for whatever reason, search technology hasn't really gotten into this kind of thing - you still can't text-search the internet archive, for example. (Can you?)
They have an active website archiving program, so they have some experience already.
60: Somebody like fa or ttaM can speak more knowledgeably than I can, but in general, acquiring everything you can get your hands is not a viable long-term approach. There are indexing needs. There's a need for *archival* storage methods (calling on ttaM and others here). There's sheer overload.
It's really not workable in the long term, especially if you know going in that an increasing percentage of what you're storing is dross on the order of "I need to floss my teeth more often" or "Fluffy's been peeing all over the house again" or whatever. A collection that becomes increasingly dominated by junk is not a collection.
62: those Long Now bastards have gotten to you, haven't they?
You know who flossed the hell out of his teeth? Nietzsche.
Nietzsche had teeth? I thought he chewed his food through acts of sheer will.
(I don't actually know anything about Nietzsche.)
And ttaM would know more about the practical aspects than I do. I've still only just read, or heard said about stuff.
66: It's possible he had trouble with his umbrella.
And you know what?! I knew someone -- probably Tweety -- would bring that up.
52: Is there a capitalization issue here? :D seems more likely.
Downloaded the whole dataset and there are only lower-case letters so I suspect either the search and/or subsequent processing combined the cases and presented them as lower. Or even other more wronger scenarios.
52: first
(rising predictably to the proffered flashy bait)
That's what makes horse races.
Any text editor that requires a mouse or menus strikes me as a hindrance.
When you first try to learn vi, it seems punishingly obscure, but it was designed by programmers who work with flat text all day every day, to serve their own needs, and once you know it, it is very very productive. And it's lightweight: starts instantly, you can open fifteen files in fifteen different vi windows without driving your system to its knees. And your fingers stay on the homerow As God Intended.
I used to be extremely fast with DEC's edt, and xedit on IBM VM/CMS has its experienced and discriminating partisans.
Some programmers I used to know and respect very much liked bbedit.
I understand why emacs is the way it is, and I have a sort of intellectual admiration-from-a-distance for its capabilities, but I have never been able to like it.
and your fingers stay on the homerow
I have no idea what vi is, but this is a significant plus for any software.
Advantage over what? It's not like there was some comparison decision making going on, as far as I know.
I'm thinking mainly of blogs. How would you archive the blogosphere?
And yeah, it doesn't look like that's what the LOC was actually thinking along those lines exactly, in terms of comparison, but I think that's largely because this sort of thing isn't even really doable with anything except Twitter at this point.
On the one hand, the LOC is in the business of archiving lived creative output. Okay, and that has been very valuable. On the other hand, it doesn't archive every conversation overheard by people walking down the street.
But does it not do that because the stuff is worthless, or because it's impossible to do?
In general, I don't think the present is a very good perspective from which to tell what information will be important in the future.
What about electronic lab notebooks?
Anyone use one? Or do you just type everything into blogger?
Personally, I think the Twitter archive is a fantastic idea.
You know, I used to be kind of down on vi (vim, really; I am but a child) and emacs, but I find myself drifting ever more often to the console for editing. I suppose this means I've become a man. Joel is right, though.
What about electronic lab notebooks?
For a while I was using a Livejournal pretty productively for this. But I haven't kept it up.
I don't think the present is a very good perspective from which to tell what information will be important in the future.
Sure, but (a) it's the only perspective we have, (b) there are people well versed in looking ahead, and back, and around, and ahead again, who build collections, and (c) what's the alternative? To archive absolutely everything?
The fact is that we cannot obsessively record everything. Not only is it not viable on a practical level, it's not the way of life as we humanly live it. Some things may be become lost that we come to regret, but we have to make the call. I don't say that it should be made on a whim.
How would you archive the blogosphere?
You wouldn't. The blogosphere and twitter aren't really comparable in this way. It's like asking "how would you archive magazines?" and comparing that to archiving a particular organization's output.
In general, I don't think the present is a very good perspective from which to tell what information will be important in the future.
Sure, but it's hard to get from this point to any particular decision about what to save. There's a relatively large literature on this, but very few guidelines everyone agrees on. It basically comes down to institutional policy. I don't think there's any line the LOC can draw that keeps twitter out and some other things they already have in; I just think they were rash.
I do think the present status of twitter is probably the main reason they jumped to acquire it. I don't think the people who oppose it or are critical of the decision are the ones judging too much by the present.
You wouldn't. The blogosphere and twitter aren't really comparable in this way. It's like asking "how would you archive magazines?" and comparing that to archiving a particular organization's output.
Right, that's what I'm saying.
Oh, and I should have added, you wouldn't archive "the blogosphere" but it's not particularly complicated to archive individual blogs.
Here's the LOC website archiving site, if anyone is interested. There are blogs there, among other things. I'm not sure that's the best way to go about it, but they've been doing it for a few years already.
I'm mostly just trying to push back on the idea that archiving Twitter isn't worthwhile because it just records the minutiae of daily life. That's the kind of thing I would say is absolutely worth saving if possible, which is usually isn't.
83: It sounded like you were saying it was a choice between comparable alternatives. I'm saying they're not even alternatives, just different categories of things.
86: No, I'm saying that if you want to archive some relatively comprehensive slice of the internet, Twitter is a good choice (really the only practical choice at this point).
71: I'm so glad joel hanes wrote something techy, because all his other posts in the last day or two had made me think he was basically me.
85: Ok, and that does seem to be what parsimon is arguing, but it's not what I'm arguing. I've got no problem with minutiae. I've got problems with open-ended public organization's commitments to keeping private organization's output for them without really knowing what they're getting into first. The LOC has a lot of stuff to do and a currently functioning organization's data may not be the highest priority compared to stuff that's getting lost or at risk. As far as I can tell, most of the costs and commitments are on the LOC side.
I suppose if I had a higher opinion of twitter as an organization, I'd be less skeptical, but I don't think their stuff is well-designed. They're successful at attracting users, so I'm probably just a cranky person without a fully internet capable phone who would like to be able to enter a real url.
I guess a better way to say it would be that Twitter's the only major part of the internet that can be comprehensively archived. The comprehensiveness is the point.
I've got problems with open-ended public organization's commitments to keeping private organization's output for them without really knowing what they're getting into first. The LOC has a lot of stuff to do and a currently functioning organization's data may not be the highest priority compared to stuff that's getting lost or at risk. As far as I can tell, most of the costs and commitments are on the LOC side.
All valid concerns, of course, and I don't know enough of the details of the deal to know if they've been adequately addressed. I don't really have any opinion on the actual LOC Twitter archive, about which I know little. I'm just in favor of the idea of archiving Twitter.
I'm skeptical of twitter's comprehensiveness. But then I'm also skeptical of thinking of it in terms of "archiving the internet."
"I have a map of the United States... actual size. It says, "Scale: 1 mile = 1 mile." I spent last summer folding it. I also have a full-size map of the world. I hardly ever unroll it. " -- Stephen Wright.
91: I guess it's mostly comity, then. But twitter is already archived, in the sense of the public archive that they may or may not have ready to transfer to the LOC. It's not really different than having an in-house corporate repository.
Now if twitter also committed to transferring their internal, institutional records (think business archive, rather than publication archives), that would be something. I'd like to see the two go together, if they're going to go anywhere. You obviously would put suitable access restrictions on the business archives side, just like with any other business records.
I probably shouldn't have brought this up here, because I was going to write something about it by my own name. Oh well.
What percent of this archive is going to be pictures of birds trying to lift a whale?
85, 90: Alright, I'll think about this. I don't have an objection to archiving minutiae. Perhaps -- perhaps! -- it's that I think twitter is artificial: in many cases, tweets are performance, like people speaking from, or as, personae. You're not getting anything remotely like an unfiltered view of life as it was in 2010. Unless you in fact are ...
I'm not sure, but I take the point about the in-theory unobjectionable nature of minutiae.
Twitter's the only major part of the internet that can be comprehensively archived
Isn't Google archiving the internet pretty comprehensively? For some definition of archiving.
I'm also skeptical of thinking of it in terms of "archiving the internet."
Me too, actually, but I suspect there are quite a few people out there who think of it that way.
I'm skeptical of twitter's comprehensiveness. But then I'm also skeptical of thinking of it in terms of "archiving the internet."
Yes, this.
But twitter is already archived, in the sense of the public archive that they may or may not have ready to transfer to the LOC.
Yeah, that's a good point, and it does make the LOC thing seem odd. It doesn't necessarily speak to Twitter's reliability as a company that they would want to do this.
Perhaps -- perhaps! -- it's that I think twitter is artificial: in many cases, tweets are performance, like people speaking from, or as, personae. You're not getting anything remotely like an unfiltered view of life as it was in 2010.
No, but so what? This is true of all historical documents.
I'm thinking of this mainly from the perspective of having a huge corpus of daily minutiae that historians in, say, two or three hundred years will have access to. You could do amazing stuff with that kind of data, provided you have the right tools to access and filter it.
100: I was heading toward suggesting that it wasn't comprehensive, but then hedged and pulled back from that, since there's an ambiguity in the way "comprehensive" might be understood here.
To the extent that Twitter is what it is, yeah, sure.
By "comprehensive" I really just mean that Twitter is essentially all one thing, owned by one company, so you can archive it as a whole (more or less) in a way that isn't really possible with any other widely used communicative medium.
And it is a widely used communicative medium, is the key thing. There's a ton of stuff there to archive.
And yeah, there is the public archive, and presumably the underlying database kept by the company, but I'm pretty skeptical that Twitter as a company is still going to exist in 300 years. I think there's a much better chance that the LOC will, though.
And it is a widely used communicative medium
I'm heading to bed shortly, but: how many people are there on the planet? How many of them use Twitter? There's a ton of stuff there, yes, but it's limited to what I suspect is a smallish percentage of the population. I'd like to see stats on that kept, then I'd probably have no beef against any use future historians might make of the data.
how many people are there on the planet?
About 6.8 billion.
How many of them use Twitter?
Over 100 million, so about 1.5%. So yeah, only a small percentage of the whole world, and disproportionately rich, no question. But again, so what?
Basically, all the problems with Twitter as an historical document (and there are many) apply equally to all other historical documents as well.
As I said, as long as the data carries a rider clarifying that this data comes from 1.5% of the population, I'm cool.
I'm certain that the LOC will be around longer than Twitter. But that still doesn't mean the commitment has to be made now. Personally, I'd like to see twitter make their public archive actually publicly available - presumably it exists, but it's really hard to get access to old tweets: at least, I've always gotten errors pretty quickly when I've tried to scroll back through search results chronologically - now. And I'd like to see an agreement from them to donate a larger share of their stuff - the business records particularly - in the future. The early reports were that they were just going to send published tweets and none of the underlying data (because it's proprietary, or because of privacy, or for other reasons that I do think are legitimate given that it's a currently functioning business, but that would not prevent a transfer under different terms of access and preservation). Your ability to analyze the twitter corpus is going to be pretty constrained if you don't have the follower/following data, for example.
Anyway, my lowish impression of the company comes from the way they make the service available. The web interface isn't particularly good and there has never seemed to be any effort to make it much better. Yes, you can get better third-party applications, but that's not the same thing.
That would make it both much more representative and much better documented than virtually any other historical document, so fine.
Your ability to analyze the twitter corpus is going to be pretty constrained if you don't have the follower/following data, for example.
Yeah, this is the sort of thing that would really need to be worked out to make any archive useful.
Twitter pretty comprehensively covers the twitter part of the internet. If you can keep the links working it takes in a bit more. But there are different ways of measuring comprehensiveness, and on some of them the cross-section captured in the internet archive's web archive, for example, is surely more comprehensive.
But archiving (broadly speaking) isn't really about saving subsets of stuff, it's about saving evidence of the activities that generated the stuff that can be saved. Take the canonical lunch tweets. Sure, they're materially part of the internet as tweets, but they're also reflections of the lunch that was eaten (if they're truthful), or of how people use the sharing of everyday life to form or cement or maintain friendships, or any number of things that happen across the internet but can't be reduced to percentage of the internet or even percentage of the population. Which is why I'm not against archiving twitter, just against doing it poorly.
Which is why I'm not against archiving twitter, just against doing it poorly.
With the caveat, already above, about costs.
So no one has any recommendation for a third-party twitter application? (We've probably bored everyone to sleep.)
I should note that I don't actually use Twitter myself. I just find it interesting as a social phenomenon.
I basically use it only for professional reasons, because people in a couple fields seem to use it as their primary non-direct-to-individuals way of communicating stuff. But I can't be online all the time, so without filters and some way of saving things I basically get only the stuff most recently posted and maybe a little more when I scroll through feeds until I get sick of the interface.
It was a hell of a lot easier to follow along from a distance when I could go to a blog, read a post, and read discussion of that post attached to that post. Following a conversation between feeds on twitter.com is really difficult. It's also annoying to get to a post and see that almost every "comment" is the test of a tweet. And then maybe there's an interesting comment in paragraphs that no one, including the post's author seems to have responded to, except maybe on twitter. I've made all of these complaints before. Maybe I should just link this comment again next time.
It was a hell of a lot easier to follow along from a distance when I could go to a blog, read a post, and read discussion of that post attached to that post. Following a conversation between feeds on twitter.com is really difficult.
Yeah, this is one of the main reasons I've never seen any reason to use Twitter myself. It does seem to be widely used in a few circles that I have reason to keep track of, so I read a few feeds from time to time, but it's really not that easy to follow the conversations so I don't make it that regular a practice.
Setting the practical issues of how to store and make useful a comprehensive Twitter archive aside, there's also the privacy question. Twitter is supposed to be epheremal, an online conversation, somethign you throw out there and which is gone the next day (or the next minute even). If every stupid remark you ever made on Twitter will be kept _forever_, that's quite a different thing. Especially if the archive does become useful and accessible to the great unwashed.
From a research point of view though I can see a complete Twitter archive would be great for datamining, the first time you can get a comprehensive snapshot of the state of the world's mind (or at least its richest 100 million people or so) at any given time.
First thing researchers need to do though is proper threading. Any conversation on Twitter more than 3 tweets long gets difficult to follow.
121: If every stupid remark you ever made on Twitter will be kept _forever_, that's quite a different thing.
Deja News redux.
||
And the "well, sure computers can do this. But that is hard" wheel keeps turnin':
Computers with artificial intelligence can be thought of as the machine equivalent of idiot savants. They can be extremely good at skills that challenge the smartest humans, playing chess like a grandmaster or answering "Jeopardy!" questions like a champion. Yet those skills are in narrow domains of knowledge. What is far harder for a computer is common-sense skills like understanding the context of language and social situations when talking -- taking turns in conversation, for example.
You thought the Jeopardy answering was pretty impressive three days ago when it was IBM instead of Microsoft you were fluffing, Times.
|>
Somebody they'll teach a computer to properly close my tags.
pars: Perhaps -- perhaps! -- it's that I think twitter is artificial: in many cases, tweets are performance, like people speaking from, or as, personae. You're not getting anything remotely like an unfiltered view of life as it was in 2010.
teo: No, but so what? This is true of all historical documents.
Gwen DeMarco: They're not ALL "historical documents." Surely, you don't think Gilligan's Island is a...
[All the Thermians moan in despair]
Mathesar: Those poor people.
We are the veldt's historical documents.
So no one has any recommendation for a third-party twitter application? (We've probably bored everyone to sleep.)
I wasn't asleep (well, for part of it I was; mostly I was without power—and still am at home; fuck you, storm!). Anyhow, I have to dabble with Twitter occasionally for work-related reasons, and the other Twitter dabblers, who dabble much, much more rave about Tweetdeck. I don't know anything about Tweetdeck, though. I just hear it raved about. A lot.
And, as to the multiple mentions of "vi" in this thread, I take as a sign of my cosmopolitan sophistication that I kept wondering: British or US Virgin Islands?
There are Spanish ones, too, actually (Vieques and Culebra), so I was being, um, oversightful.
Ok, I'm giving tweetdeck a try. It seems to be for more intense users, but hopefully I can turn a lot of features off.
If US going to lose, much happier to have it be someone like Ghana than many others.