Update on the Twitter Archive at the Library of Congress | Hacker News Reader

Update on the Twitter Archive at the Library of Congress

blogs.loc.gov

59 pointsdizzystar42 comments

I'm so old I can remember when you could see all the live tweets flying by last decade.

Great quote in the (too long) nyer article: “Talk, talk, talk: the utter and heartbreaking stupidity of words,” William Faulkner 1927..

I'm sure the NSA, GCHQ, the big platform companies etc are archiving all the tweets forever, it's a shame the public don't have access to curated content taxpayers have paid to keep in those repositories.

The LoB can't possibly archive the entire internet without massive investment

I've wondered how much hardware it takes to store and search over every Tweet. I figure the total number of Tweets is in the low trillions. I know Twitter does this themselves and presumably http://support.gnip.com/apis/search_full_archive_api/ uses the same backend servers. Does Twitter have any public info on how much hardware is dedicated to this?

It's sometimes mentioned in their blog posts:

https://blog.twitter.com/engineering/en_us/topics/infrastruc...

Not a lot of specifics, but there's some interesting tidbits:

> Hadoop: We have multiple clusters storing over 500 PB divided in four groups (real time, processing, data warehouse and cold storage). Our biggest cluster is over 10k nodes. We run 150k applications and launch 130M containers per day.

> the cost of properly arranging and organizing vast amounts of information is frequently underestimated.

There are a lot of jobs to be had here. But not in tweets. Think cancer research, molecular biology, taxonomy, behavioral and experimental psychology. The data

> There are a lot of jobs to be had here. But not in tweets. Think cancer research, molecular biology, taxonomy, behavioral and experimental psychology. The data

Did your comment get truncated?

He stopped at the 160-character limit to make a point.

Thank you; it totally went over my head.

It seems like a good decision. I don't think there's much merit in archiving the entire firehose.

It's impossible to know ahead of time which tweets or trends are going to be interesting.

True, but filling the records with so much noise that finding the interesting ones amounts to concentrating gold from seawater, you’re no longer doing the job an archivist should.

If you have the firehose, it is easy to derive the selective archive by filtering on the list of noteworthy accounts. Ignoring monetary costs, archiving the firehose is strictly better than selective archival. It enables future researchers to examine trends on a societal scale.

Or find things that seemed insignificant at the time but became significant later with hindsight.

not to mention the fact that twitter is basically millions of monkeys banging away at keyboards, eventually they will produce some great works.

For values of “eventually” which include timescales on the order of proton decay...

...like a particular painter and WW1 veteran in Germany that the than current art world dismissed...

It does, at the cost of preserving the firehose, through endless format and media shifts, and at the cost and under the assumption that future historians will want an archive they need a strong AI to manage. There is also the cost, as in paying people and buying equipment, drawn from an already overextended budget.

Frankly, the firehose will be considered priceless in 100 years. The accounts and tweets that we care about now will also probably be the least valuable. Historians and archaeologists will generally take one grocery list over a thousand sweet nothings from politicians and celebrities.

Preserving minutiae for future archaeologists is a bold and exciting idea which should be funded and pursued. It is not however, the mission of the LoC.

Edit: not to mention that I think you have a far too rosy view of the world we’ll be leaving for people 100 years from now.

Agreed. Hell, I've recovered old hard drive backups from ~10 years ago and have been fascinated by by what I've found. Old e-mails and chat logs and stuff can be very entertaining to dig through.

There's so much valuable research that can be done with aggregate Tweets that you can't do with just lone accounts. For example I've seen flu outbreaks being predicted by hashtags. (http://money.cnn.com/2017/05/11/technology/tracking-flu-twit...) The possibilities are really endless, but depend on whether you have the data and whether it's over a large enough timescale for your use. Twitter represents a large portion of our culture and just because you don't personally like it or have a use for it, doesn't mean we should just throw away archives of it.

Given that tweets are only 280 characters and there are perhaps a couple hundred accounts that the LoC archives, I fail to understand why the premier archiving project in the country can't keep a slowly growing fistful of gigabytes on a hard drive somewhere.

Agreed. I have to wonder if they are doing it manually.

From someone who worked with the Twitter firehose briefly: A handful of gigs per day. More like hundreds of terabytes of information total, headed into petabyte territory. First, there's a lot of tweets. Also, the message itself is not all the data available with a tweet. Toss in any semantic processing or tagging and each message grows quickly.

The company I worked with spent a non-trivial amount of money storing historical tweets. I'd even go so far to say that was the majority of the IAAS costs - even more than the compute required to process them in real time.

OK fair enough, I didn't account for metadata. I'm gonna try some back of the envelope math, tell me if I go off the rails anywhere:

First, how many accounts are we talking about here? Between POTUS, VP, their spokespeople, cabinet members and official agency accounts, I'm going to assume the executive branch has about 50 accounts that ought to be archived. For congress, I think it's reasonable to archive each member's account and their spokesperson: 200 accounts for senate, 870 accounts for congress. In the interest of Fermi estimation (and because I'm not sure if every one of these people has an account) let's call it 1000 accounts.

I'm going to go with a mean of 10 tweets per day (again in the name of Fermi estimation).

With 280 characters + metadata, I'm going to round up to 1KB per tweet.

1000 accounts * 10 tweets/day * 1KB/tweet = ~10MB/day = ~3.65GB/year = ~40GB for the current lifetime of Twitter

If you're drinking from the firehose to archive tweets for a huge userbase (and feed them into models or perform semantic analysis) I could see this getting expensive and costly. If you're just try to archive tweets from a certain group of users and keep them on a disk (with a tarball or zip file released quarterly) it feels a bit more doable.

That said, this little thought exercise has gotten me thinking a lot more about what I expect of the LoC and the National Archives. I'd be happy with a "cold storage" record of the tweets being preserved for posterity, but they may see their mission as making the tweets into a tagged searchable collection. I also think there are arguments to be had about how many people's account really need to be archived (perhaps states could handle archiving their own reps).

In the end I guess I'm cool with the LoC scaling back collection as long as they're transparent about how they're doing it.

They were doing all accounts previously, not just government officials. Their new selection is likely to still include government officials.

It's not just metadata.

You can tweet photos, videos and GIFS. My average gif is 500kb.

If these were excluded I suspect it would make the archive much easier to manage.

Any archival project worth its salt would never store just the tweet text, but also all associated metadata. The best way I've seen to do this is to just save the JSON returned by an API as it's easily parseable. You also have to consider that the utility of Twitter isn't in the single tweets, but rather the whole comment chains. Lastly, Twitter has images as well.

> there are perhaps a couple hundred accounts that the LoC archives

The article states that it was archiving _all_ of public twitter. It is moving to the selectivity that you have assumed was status quo.

All of twitter, simplz gz compressing the firehose stream, is at least 4tb/month. Likely more

I feel I'm not the only one thinking that "select few" to archive will exclude a certain public figure. The benefit to the "full dump" is that it includes everything, without any bias or removals.

Are you implying that the Library of Congress will exclude communications from the President of the United States? That seems like a stretch.

The article title is misleading, they haven't "quit twitter", they're just no longer archiving every public tweet (except for a select few, e.g. those that are likely to be historically interesting/noteworthy - e.g. Trump's twitter utterances).

Perhaps the title should be editorialised to something like:

"Library of Congress announce a change in collections practice for Twitter"

edit:

@dang - how about changing the link to:

https://blogs.loc.gov/loc/2017/12/update-on-the-twitter-arch...

https://blogs.loc.gov/loc/files/2017/12/2017dec_twitter_whit...

Misleading, maybe, but I don't think many were confused about the actual meaning of the headline (I certainly wasn't).

So you actually read it as "The library of congress quits archiving Twitter"? Hmm...

Yes. But I had heard of the news story about them starting to archive twitter. If I hadn't, maybe I would be more confused. Why would the library of congress quit twitter, I thought, and why would it even be a worthy enough event to warrant a news article? If they didn't find value in it, they would just stop posting. Why would a neutral government agency take something like a political stance against twitter? If I was wrong, yes, it definitely was news worthy. But occam's razor, and the law of clickbait, led me to the conclusion that most likely, it means that they've stopped archiving tweets, and not that they have deleted their account.

Hmmm...I dunno, I initially wondered if the current White House residents had issued a diktat shutting down their twitter account. Not an unimaginable thing in these strange times.

I and probably many others read it as the twitter account @librarycongress being deactivated.

The tone of the article seems also rather biased against Twitter. For example, it mentions;

> “Today, we announce a change in collections practice for Twitter. Effective Jan. 1, 2018, the Library will acquire tweets on a selective basis—similar to our collections of web sites.” The phrasing was elegant, but the sentiment was nonetheless familiar: “Quitting this shit!!!!”

No. That's not at all the sentiment I get from that. The volume has probably grown too big and it makes little sense to archive spend money archiving all of it.

Yep, the New Yorker article seems to be trying to make a fuss out of something that isn't there.

These days I find the relationship between writers and twitter more interesting than twitter itself. It's like they still haven't internalized the properties of the medium that were well-understood back in 2010.

Maybe they're just the voice of perspectives that are more widespread, but every news or pop culture take I see about twitter seems to be born from the unexpectedly-complicated relationship they seem to have with becoming yet another head in an unimaginably large crowd of people. Where everyone can talk to everyone, anyone, or nobody in particular.

I agree. URL changed from https://www.newyorker.com/culture/cultural-comment/the-libra..., which points to this.

Appreciated.

Reply on news.ycombinator.com