Slashdot Mirror
Snapshotting the Whole Internet?
Anonymous Coward writes "CNN is running a story about a company that is saving periodic 'snapshot' archives of the whole www (or as much as they can) for historical purposes. Interestingly they say that although they might have considered saving everything except ads, they didn't throw away the ads because historians claim that ads give a better "glimpse of what life was like" in the past. I wonder what legal ramifications will arise for possessing such archives of the "whole web" as snapshots-in-time. Thoughts of DeCSS, CPHack, MS Kerberos' click-wrap license, I.P. "ownership" of collected databases cross my mind."
hm. my hosts file is preventing me from easily reading the whole CNN article, but here's an article about a possibly different company doing the same thing, dating back to 97:, the website itself is http://www.archive.org/.
http://Slate.msn.com/webhead/97-02-27/webhead.asp
The related Xerox project I think is merely affiliated with Archive.org, actually, and is currently called the Internet Ecologies project
Returned Peace Corps IT Volunteer
Why do I just NOT believe this? In order to do this, they would not even be able to search for httpd at every IP, because it would only grab one host, and virtual hosts would be skipped.
They would actually have to search every registered domain name, dig up all it's host names, and then search each one for an httpd, and I don't believe they are actually going to do that (can someone point out to me that they are?).
One of the biggest problems on the web right now is the lack of orgninzation of information... "You want to know what? Oh, it's on the web, no doubt, but WHERE!" As brillant as search engines are, they still don't know where everything is, and you frequently will miss what you REALLY want.
So along come theses people, and they make an even larger claim than searching through _all_ of the web, but they say they will take a snapshot of ALL of it?
/me looks at hit logs for home Linux box running apache on modem... Hmm... Don't see anything... been up weeks... I'll keep looking ;-)
At best, they will get the most popular sites, and try to leave it at that.
LynxInformation
/=search[delete]=historylist
Lynx
Lynxisa textbrowserfortheWorld WideWeb. Lynx2.8.3runson
Un*x,VMS, Windows95/98/NTbutnot3.1 or3.11,onDOS(386 or
higher)andOS/2 EMX.Thecurrent developmentalversionis also
availablefor testing.PortstoMac areinbetatest.
*How togetLynx,andmuch moreinformation,is availableatLynx
links.
*Many userquestionsare answeredintheonline helpprovidedwith
Lynx.Pressthe'?'key tofindthishelp.
*If youareencountering difficultywithLynxyoumay writeto
lynx-dev@sig.net.Beas detailedasyoucanabout theURLwhere
youwereontheWeb whenyouhadtrouble,what youdid,whatLynx
versionyouhave(try'=' key),andwhatOSyou have.Ifyouare
usinganolderversion, youmaywellneedto upgrade.
Maintainedby lynxdev@browser.org.
Commands:Usearrowkeys tomove,'?'forhelp, 'q'toquit,' Arrowkeys:Up andDowntomove. Righttofollowalink; Lefttogoback.
H)elpO)ptionsP)rint G)oM)ainscreenQ)uit
_______
2B1ASK1
i parsed this as
"Slashdotting the whole internet"
help me....
Now after Microsoft goes on their Jihad against warezers trying to take down all illegal copies of their programs, you can consult these archives to find some choice applications.
A deep unwavering belief is a sure sign you're missing something...
That was me. I just forgot that after previewing, you lose your login if you don't have cookies on.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Joey: Wow Grandpa, this was how the Internet was back in your day?
Grandpa: Yep Joey, that's right, this archive shows how the Internet looked back when I was a youngster like yourself.
Joey: Wow! You know Grandpa, it's too bad what the Internet has become now. Just look at this...porno, warez kiddies, AOLers, and ads everywhere! Talk about a downward spiral.
Grandpa: Joey?
Joey: Yeah Grandpa?
Grandpa:: You're still looking at the archive.
--
--
The real Raunchola isn't cool enough to have any imposters
. . the old internet before micrsoft.
___
Sounds like an excuse for having the worlds largest pr0n collection 8^)
.
Future historians will no doubt be fascinated by the "ancient mystery" race of trolls. They developed their own hybrid latin/arabic based alphabet, but often reverted to more primitive picture-based communication. The figurehead of their religion was a deity named Natalie Portman, who they worshipped with sacramental rituals involving hot breakfast cereals. Who these people were, and how they really lived, will sadly remain a mystery.
There is info on the side on how the archive is accessed, created, who pays for it, everything. Read it before you hit that post button another time.
---
END OF LINE
It's not a question of converting anything into a legal document: it's a useful source of additional evidence, though, if the owners let it be used that way. Short of new legislation, though, they're under no positive duty to help someone trying to sue a site owner.
In practice (and this goes for more or less all of the legal world) you can get evidence of what was at a given site by traditional methods with only a little bit of technology. It's not outside the rules of evidence for a witness to stand up in court and say, on oath, that he saw such-and-such on a site one day.
Indeed, that's how nearly all documentary evidence goes in unless both sides agree it's kosher: someone has to stand up, swear in as a witness and say on oath that a particular document came into existence in the form it is in before the court on the date written at the top, etc..
Producing a hard-copy of the offending page with someone to swear he downloaded it on the date shown at the bottom and printed it out would be quite enough, without having to go pester someone who wasn't a party to the action.
And this is before you get discovery of the other side's HDD and run some undeleting software on it. That's been done in a couple of decided cases and certainly the UK courts are happy with it provided the party doing it wheels a geek in as expert witness (and you guys are among the cheapest expert witnesses going: sharpen up, OK?) to reassure the judge there's no jiggery-pokery involved. I've got a case on at the moment where we've done just that.
-- AndrewD
A Maze of Twisty Little Laws, All Different.
let me try and clear this up for everyone: JE is a troll.
But what's worse is his strategy of arguing against humanities' overall increase in knowledge, and having many of the lurkers take this stuff seriously. Those lurkers in turn become the grousing bitches in the workplace, the fux0rs that make it impossible to achieve greatness. Daikatana is a perfect example of not only inept management, but a bunch of nasty grunts that regurgitate base cynicism, by way of sucking up this sort of troll born tripe.
Why would you discourage the archiving of such a valuable chunk of data? Besides just eliciting responses... In the final analysis, there is no reason not to support (or at least stand clear of) these archive.org folks, is there?
In short, this is the crank-assed behavior that hobbles the entire high tech world. This is not a well formed criticism or a burst of insight, this is a wanky know-it-all piece of crap designed to spark a useless debate. The problem is, that is more damaging than anything else going on on the web today.
Remember 1994? Monica Lewinski was just another intern. peecees were still 16 Bit. Linux was 1.0 and a guy named Jim Clarke started Netscape Comunications.
___
The "teething problems" are what makes it historically interesting! I think historians of the future will be much more interested in looking at the development of the web through trial and error than at the finished product.
Well I suppose that the sheer amount of perversion and degredation available on the net at this point in history will provide a lot of interest to future historians, so in that context sure it'll be "historically interesting"!
But, pornography aside, what is there of real historical value on the net? Sure there are any number of mindless geocities homepages full of drivel about people's pets, but sifitng through this would drive anyone mad and there are a lot more "insightful" sources already available about today's culture.
Unfortunately the web as it stands at the moment shows the worst side of humanity rather than its best side - historians looking through terabytes of things like the anarchists cookbook, virulent anti-Christian diabtribes, terrorist manifestos and race hate sites will hardly pick up a balanced view of society will they?!
It's not a study, it's an archive. The purpose of this project is to collect data, not to analyze it or place any sort of value judgement on it.
But unless it will be used as the basis for future studies then this project is a waste of time, so I don't think you have a valid point here.
---
Jon E. Erikson
Jon Erikson, IT guru
Alright, chances are the growth of the Internet (and particularly the web) will eventually be recognized as one of the major sociological developments of the late 20th century. Someday, we'll want to look back at the roots of the revolution and trace its development through the present day. However, that may well be an impossibility: we still have Gutenburg Bibles, but the original Mosaic/Netscape site is already a dim memory.
The other day, I was browsing through the Hotwired archives. Basically, they have everything to come out of the Wired family of publications for the past five or six years, and that's great: it's fun and oddly fascinating to read an article from early '96 and hear about this fantastic new "push" technology. But, being a Wired venture, many of the stories are gaudily hyperlinked, and very, very few (if any) of those external pages are still extant. Entire dimension to these stories are already lost to the ages.
The are a lot of obvious difficulties in archiving the web, but it's something that probably should be done. I really think not too far down the line, we'll look back and regret that a lot of what we take for granted today wasn't preserved.
-jay
Our website is at http://www.archive.org.
We are *NOT* a company. We are a non-profit organization making our archives freely available to researchers, scholars, historians, etc.. A for-profit company may not be the right model to insure long term longevity of the collections. We only archive publicly available information on the Internet.
We currently have about 17TB of Web pages and images on disk. We've also got about 6TB of older stuff on tape that we are migrating to disk. We're growing at about 3-4TB/month. We are not yet getting Usenet or streaming media because of labor limitations. Anyone wanna come work for us?
We buy storage PC's with twenty 75GB IDE hard drives, 2 667Mhz CPUs and 512MB RAM. We run Linux, but are migrating to FreeBSD because of the 2GB file size barrier.
Access currently requires a bit of UNIX skill. There is no browser interface to our collections. You'll need to be able to write your own search software, as the only index we have right now is a URL index. If you want access, you'll need to fill out a form at http://www.archive.org/proposal.html.
Kurt Bollacker
Technical Director, Internet Archive
kurt@archive.org -- www.archive.org
P.O. BOX 29244, San Francisco, CA 94129
vox: 415-561-6796 -- fax: 415-561-6768
Prevent email address forgery. Publish SPF records for y
Another group, the private company Alexa, has also engaged in internet archiving. A couple of years ago Alexa donated an archive to the Library of Congress. It was written up in Brill's Content last November (article text not online, alas).
(Alexa's normal business involves a browser plug-in that is What's Related on steroids.)
----
lake effect weblog
{Network engineer in Chicago--looking for work!}
Jon Erickson spewed: ... any study that attempts to categorise how we live at the moment using the web is doomed to be prejudiced and incomplete. Until everyone is online and has equal access, this is just another arrogant study attempting to categorise who is worth enough to be able to use the net.
[much repetition deleted]
So what you're saying is that the ONLY records worthwhile to historians are those that reflect "everyone"? Just who the hell are you to tell future historians what will and won't be useful to them? Do you understand history at all?
The importance of any given historical document is often not found in the document itself, but in that document's context.
My father managed a professional local historical society and museum while I was a kid. The museum was a high Italianate mansion in the upper Midwest built by a lawyer who got rich on land speculation. You are correct that the biggest danger facing future generations is to look at a museum such as a rich lawyer's house and believe This is the way people lived in 1870. In fact, only about 1% of the people lived that way (servants, nice furniture from Europe, that sort of thing). To help counter this one of the first things my father had done was to take custody of a "fortuitously" threatened building, one of the first in town, which was more typical of your average family.
By themselves, neither building is representative. Even together they fail to represent everyone in this particular community. But by putting them together in context you can illuminate things that would otherwise be much more difficult to see.
What a snapshot of the web ca. 1996 shows is most definitely a subset of the larger society. But you can't say to me that it doesn't say anything useful. You yourself note that the people, ideas, and connections that it shows are a particular and identifiable subset of the larger society. Realize then: That, in and of itself, is useful to historians. Even more so is the randomness and complexity of the archive, because having a human select what future generations of historians will find useful or interesting is a dubious proposition.
If you're waiting for the day when "everyone is online" to start recording our digital history, you'll have to hold your breath a very long time.
----
lake effect weblog
{Network engineer in Chicago--looking for work!}
I don't know if this has been touched on yet, but it seems to me that the archive would make an ideal 'memory' for an AI to grow in to.
During the AI's maturation cycle, one could simply provide the data raw and let the pattern matching and cross indexing fly.
The archive is mixed with every aspect of the human mind, the art and music, the filth and evil, the self-righteous religious fury, the meek and mild post-hippy sentiment, the peircing precision of scienctific method and the wacky good fun of psuedo-science madness. The full spectrum of humanity is presented here, from autistic to genius, idiotic to brilliant. There are plenty of wholesome sites, children friendly basic learning resources, museums and FAQ's, on and on.
I propose that the archive be used as the base for intuition in a full scale AI project, with the actual info rarely making its way up the stack to conciousness, instead providing the hints and connections necessary for a model of human memory. Explicit data filtering will be required for the prototype and experimental versions, but a truly 'human' AI would need to understand humanity on some pretty deep levels if we were to expect it to relate to us, and have sympathy/empathy for us regardless of our shortcomings.
Maybe this process is already happening under the guise of the internet search engines!!! There is an awful lot of pressure and incentive to provide relavant responses... what better way to accomplish this, than to have some entity think about the list of matches and understand human requests like a human?
:)Fudboy
I guess I'm just a Fudboy, looking for that real Transmeta...
:)Fudboy
I guess I'm only a Fudboy, looking for that real Transmeta
I wish I hadn't just used my last moderator point. ...</I>
...</I>
...</I>
<BR>
<BR>
<I>Sure, this'll be a useful reference for future
You're partially correct. Yes, the web is in a "teething" phase. Yup, glitches do tend to override content, in some places. But it's still worthy to archive. How will future generations answer the questions: How exactly did the Internet start out? What did it look like?
With these snapshots, of course.
<I>The trouble
I feel this is pretty much completely correct.
<I>So, given this
I don't think it is getting a snapshot of how society is today. But is that really the point? Or, rather, is that really what these archives will be used for in the future? I doubt it. But this archive will provide a great deal of information about the growth of communications techology, and how it will change our lives.
<I>expressing views that people consider "outdated" or "primitive", even they are held by many others<I>
If a million people say something that is factually wrong, it's still wrong. Got that?
<I> Anyway, any study that attempts to categorise how we live at the moment using the web is doomed to be prejudiced and incomplete. Until everyone is online and has equal access, this is just another arrogant study attempting to categorise who is worth enough to be able to use the net.</I>
Yeah, so what? Does that mean they should stop? It's still their time and money. And this archive WILL provide an amazing picture of the Internet when it was still relatively young(teething, as it were).
Dave
Barclay family motto:
Aut agere aut mori.
(Either action or death.)
But, pornography aside, what is there of real historical value on the net? Sure there are any number of mindless geocities homepages full of drivel about people's pets, but sifitng through this would drive anyone mad and there are a lot more "insightful" sources already available about today's culture.
Unfortunately, our knowledge of history already has numerous gaps where things that, at the time, were thought to not any value. Leave the interpretation of value to future historians, and meanwhile, let's not make decisions about what isn't important enough to save.
Unfortunately the web as it stands at the moment shows the worst side of humanity rather than its best side - historians looking through terabytes of things like the anarchists cookbook, virulent anti-Christian diabtribes, terrorist manifestos and race hate sites will hardly pick up a balanced view of society will they?!
I am amazed if you think racism and hatred content is the overwhelming majority of the Internet. But that is completely irrelevant; even if this is true, embarressment of it is no good reason to censor historical records.
But unless it will be used as the basis for future studies then this project is a waste of time, so I don't think you have a valid point here.
Why do you think it won't? Why do you presume this information is, and always will be, useless trash? It is only by archiving without bias and censor that there can ever be an accurate historical record. Archive it all and let the historians sort it out.
So you're saying that because the web isn't perfect and it doesn't reflect the general society, it won't be useful to historians? If you ask me, this would make it more interesting, not less. This transition will have an extremely short lifespan (probably under 20 years in length), so the more data the better (for the historians).
And, FYI, just because the Royal Family doesn't reflect English society, it does not mean that historians don't find them interesting.
Well I suppose that the sheer amount of perversion and degredation available on the net at this point in history will provide a lot of interest to future historians, so in that context sure it'll be "historically interesting"!
You just don't get it, do you? Should historians gloss over the Holocaust, the Reconstruction, and the Dark Ages simply because they were "icky?" Sometimes the darker elements of society are the most worth examining in a historical context. The whole point of the saying about those who don't study history are doomed to repeat it isn't that you should study only the good points and avoid them.
But, pornography aside, what is there of real historical value on the net? Sure there are any number of mindless geocities homepages full of drivel about people's pets, but sifitng through this would drive anyone mad and there are a lot more "insightful" sources already available about today's culture.
Do you think it's not just as frustrating to shuffle through archives of old 19th century newspapers to find ads and articles about the medicine of the day? The point that the man speaking for the Internet Archive was making is that this is not a study of only the famous. With these archives at hand, you can study the transition from the early days of research papers to the rise of pornography and personal websites to the current days of e-commerce to whatever major social trend the web next holds. An archive of the web shows how society has adapted to the format. You can see what issues were hot enough to spur crops of websites only to fade away in the span of a year or two.
Face the music that the majority of humanity isn't putting out "insightful" commentary. Ignoring the common man is a mistake that many historians simply can't ignore because there's nothing available about them. All the "mindless" Geocities sites give an insight into the kind of people that use them.
Unfortunately the web as it stands at the moment shows the worst side of humanity rather than its best side - historians looking through terabytes of things like the anarchists cookbook, virulent anti-Christian diabtribes, terrorist manifestos and race hate sites will hardly pick up a balanced view of society will they?!
Sounds like you're the one with the hardly balanced view of society if you honestly think that is what the majority of the web is. The fact is that the majority of the web currently is commercial sites and those "mindless" Geocities sites you like to talk down about. Though there are some bad elements on the web, it's also worth historical note that the web led to the coming out of many of these fringe groups. The very anarchy and rebellion of the web is of major historical interest, and the web is becoming one of the more important socio-economic influences of the turn of the century, at least in America.
But unless it will be used as the basis for future studies then this project is a waste of time, so I don't think you have a valid point here.
Ah, but it will be. Say in 30 years you want to do some research on the Y2K histeria of the turn of the century. While there will be plenty of books to read through, a major factor in spreading the word about Y2K was the Web. However, these web sites are already mostly gone from the Web today. Fortunately, the Internet Archive may have already preserved them for future study.
Would you like to study the rise of Linux or of the web itself? Many of the early web pages about the topics could provide priceless research. Hell, even if you really object to the large amount of pornography, the booming porn industry on the web was a major driving factor in advances in e-commerce. It would also be valuable in studying the "warez" counter-culture of today.
Plus, like it or not, it's not for you to say. This is being done by a privately funded group. If you really feel so strongly that the web is worthless and should absolutely not be archived for historical purposes, then go torch the place. While you're at it, go ahead and start burning those libraries that hold material about history you object to. Otherwise, your choices are "shut up" and "like it."
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
I collect software and giveaways from ISP's, especially Win 3.1 software, tshirts, ads, old magazine articles, etc.
My facination is on the number of ISP's that emerged in the past five years and how many of them disappeared, especially the high-profile, nationally advertised companies - gnn, pipeline, etc.
I would like to see some of their archived sites from 1996 for different ISP's, their service offerings, pricing.
Call me wierd, but I find it facinating. When do we get access?
We buy PC's with 20 75GB IDE hard drives, paying about $11/GB for storage. Pretty cheap these days. We've calculated that the growth of the Web and the growth of disk drives tend to track pretty closely, so the cost of keeping up with the Internet will mean a relatively constant spending rate.
Kurt Bollacker,
Technical Director, Internet Archive
kurt@archive.org
www.archive.org
wasn't Xerox PARC already doing this? well, capturing all the text, storing terabytes at a time? I can't seem to find the link currently, unfortunately.
Returned Peace Corps IT Volunteer
Where are they going to get the capacity from? AFAIK the net has millions if not a billion sites and to store such a large no of sites would take a horribly large no of raid arrays. Also the net is dynamic and always changing . By the time they finish taking napshot of the net half the sites would have changed and btw what frequency is the best? This seems like an Augean Stable kind of task
**Life is too short to be serious**
I didn't read the CNN story closely enough. Apparently it's the same group, after all, but they are now operating under a different name for the nonprofit archive activity (as opposed to the commercial search-engine activity).
----
lake effect weblog
{Network engineer in Chicago--looking for work!}
I heard there was a version distributed over millions of servers worldwide. :P
The goal is to get all publically available Web pages and their images. Technical and labor limitations prevent this from happening as yet, but we are working on it.
How much storage space is required for the whole web?
No one knows for sure, but the best estimates I've seen put it at about 20-30TB.
What software/OSes are they using for this project?
We've got got our own software running on Linux, although we are migrating to FreeBSD because of the 2GB file size limit. As a shameless plug, we are hiring!!
When do we get to see the archive?
We intend to provide access to researchers, scholars, historians, etc. Research proposals can be submitted here.
Kurt Bollacker
Technical Director, Internet Archive
kurt@archive.org
Well, what else is it going to be used for? Your suggestion that it be used as a reference on "the growth of communications techology" is rediculous - the growth of hate material and pornography on the web has no correlation with the growth of communications technology at all. This project is not getting a snapshot of web technology, it is getting a snapshot of web content, something entirely different.
Much of the content of the web relates to the growth of communications technology. You are limiting yourself severely if you are only thinking of the raw bandwidth connections. The growth of use of non-textual content, multimedia, and scripting languages and applets are all advances in communications technology. Not to mention, the radical growth of the internet, in terms of number of sites and content on sites help document that. Plus, your view of what the content of the web actually is is rather stilted. Even if it wasn't, the vast amount of negative material is worth studying in it's own right.
Please demonstrate how believing in God and decent Christian morality is "factually wrong". I doubt you can.
<offtopicrant>
Please show how your condenscending and arrogant attitude reflects a life led by Christ. It's jerks like you who give the rest of us a bad name. Did he anywhere indicate that he was talking about Christiantiy? Anywhere? Or was he just responding to your blanket assertion that you were persecuted for your views, which in no place specified Christianity.
Instead, you assumed he did, or were attempting to sidetrack the issue to make yourself look like the oppressed religious minority. This kind of behavior is what disgusts others. When people look at me, they see someone like you -- an arrogant, bigoted ass who sees the entire world as filled with evil sinners out to get them. It makes me sick.
</offtopicrant>
My point is, if you read my post, that this is not a good thing given the exlusive access to the net by a certain portion of society. Would you consider how a society lived through records of its nobility?
Um.. Let me think. YES!! That's how historians have had to do it for ages. Should we ignore early American politics because it too was primarly run by white, middle-aged landowners? You must attempt to look at all elements of a society to see how the framework fits together. If the web is a rich WASP playground, like you assert, then it's worth studying why exactly this is. Just because one particular class was behind a major force of societal change doesn't make it not worth studying. This kind of PC "1984" style of thought would have us ignore all of our history for the goals of delluding ourselves into thinking we're perfect. Well, we're not. Get over it, and start trying to figure out why. Just because America was led by white landowners early on doesn't mean we should've ignored history to know where we are now. Similarly, we should not ignore the current Internet so that future generations will know how they got to where they will be.
P.S. Sort your HTML tags out.
Finally, the prima facie evidence of a troll. Someone you picks at the formatting rather than the content of the person they disagree with. The guy obviously accidentally selected the "Extrans" option, which I too have accidentally done before in the past thinking that it would do the opposite. Besides, you should really preview and check your spelling before being so harsh. Talk about the pot calling the kettle black.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
I wish I hadn't just used my last moderator point.
Why? Because you, like so many /. moderators, moderate anything you disagree with down? IMHO that is the biggest simgle problem with /. at the moment, with troll metamoderation the second.
I don't think it is getting a snapshot of how society is today. But is that really the point? Or, rather, is that really what these archives will be used for in the future?
Well, what else is it going to be used for? Your suggestion that it be used as a reference on "the growth of communications techology" is rediculous - the growth of hate material and pornography on the web has no correlation with the growth of communications technology at all. This project is not getting a snapshot of web technology, it is getting a snapshot of web content, something entirely different.
If a million people say something that is factually wrong, it's still wrong. Got that?
Please demonstrate how believing in God and decent Christian morality is "factually wrong". I doubt you can.
Yeah, so what? Does that mean they should stop? It's still their time and money. And this archive WILL provide an amazing picture of the Internet when it was still relatively young(teething, as it were).
My point is, if you read my post, that this is not a good thing given the exlusive access to the net by a certain portion of society. Would you consider how a society lived through records of its nobility?
P.S. Sort your HTML tags out.
---
Jon E. Erikson
Jon Erikson, IT guru
"Archiving the net is like washing toilet paper!"
Oh, go on, check out my job.
Well, of course it's not perfect. In the news story itself, they mention that sites can block the archive entirely by blocking 'bots. Also, I doubt everytime it hits MP3.com it downloads every single song. However, the attempt is being made, and from the sound of the size of their archives, they're doing a darned good job of it. Also, these snapshots obviously can't be done in just a week or less. I'm guess each snapshot takes months. Bandwidth plus the time in writing to tape archive limits how fast they can do it. They probably miss all the little dynamic IP sites too. Oh well, at least an attempt is being made. As I pointed out in another post, hopefully they've managed to capture a bit of the Y2K histeria before all the sites were pulled down in embarrassment.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Sure, this'll be a useful reference for future generations, won't it? I'm sorry, but as much of a fan of the web as I am, I really wouldn't consider it to be something worthy of archival in the state that it is at the moment. Why? Well, because currently the web is still in the transitional period between the days of ARPAnet and purely academic use and acceptance as a medium through which the general public can communicate. And as such, it's still in a state where teething problems overwhelm content.
The trouble with the web is that although it is supposedly accessable to anyone with a phone line and a PC, the harsh reality is that cost and communications infrastructure have meant that only those of a certain socioeconomic group are currently able to use the web, and this group is mainly comprised of the priviliged, a group which most /.ers fall into by dint of their jobs or backgrounds. Research carried out my both my consultancy group and others all indicate that the majority of people able to use the web are white, middle-class and certainly in higher than average tax brackets.
So given this, how does taking a snapshot of the web give a view of how society is at the moment? It doesn't, any more than looking at the Royal family of England gives a picture of what England is like (despite what some Americans seem to think). The views that are expressed on the web are those of a priviliged class who do not have to suffer the effects of current liberal free-market policies and the increasing divide between the rich and the poor.
No, this exercise will be a "Who's Who" of society, showing only those who are rich enough to be able to afford net access. The majority of people, unable to benefit from the web, will be left by this study as an underclass, something which I view as incredibly wrong and an example of the undeniable arrogance that most people on the net display towards those that are perceived as their inferiors. Indeed, I have suffered the same myself here on this forum for expressing views that people consider "outdated" or "primitive", even they are held by many others.
Anyway, any study that attempts to categorise how we live at the moment using the web is doomed to be prejudiced and incomplete. Until everyone is online and has equal access, this is just another arrogant study attempting to categorise who is worth enough to be able to use the net.
---
Jon E. Erikson
Jon Erikson, IT guru
To collect snapshots of the internet reminds me of the Wim Wenders' movie Lisbon Story. It's about a director that decides to videotape everything he can, leaving operating cameras in parks, walking around with a camera all the time, etc. He would afterwards destroy the tapes (or lock them somewhere, I don't quite remember) to avoid people to see those particular images. Some idealistic freedom for images, not to be ever seen.
I don't think a huge unfiltered database with everybody's day-to-day life has a lot of historical meaning. And I think that's what a snapshot of internet is like.
Who knows, maybe these guys are just planning to burn everything, afterwards...
-- Aber
According to this article in Nature, 1000 terabytes = 1 petabyte, and 1000 petabytes = 1 exabyte. The article notes that as ever larger and more complex scientific experiments produce ever larger quantities of data, there was briefly a possibility that we would run out of words to describe the amount of data produced. Consider that while the Library of Congress contains less than 12 terabytes, the Large Hadron Collider at the European Laboratory for Particle Physics (CERN) is expected to produce 100 petabytes (i.e., 100,000 terabytes) of stored data in 15 years... Anyway, moving on up, after exabyte, they (whoever they are) started naming things from the back of the alphabet. Thus 1000 exabytes = 1 zettabyte, and 1000 zettabytes = 1 yottabyte. Although the article does not say, perhaps the term for 1000 yottabytes will skip over 'x,' as we already have exabyte, and go to 'wottabyte'? I like the sound of that :) ...
This is really a cool idea. Up 'til know, we've taken it for granted that our media would last long enough for historians to make use of it in the future. With the web, you can't assume that's the case, so it's good that someone's taking it upon themselves to archive the web.
/. story put the number of Apache-served websites at about 10 million. Since Apache has roughly two-thirds of the market, that makes the total number of web sites 15 million. If sites average, say, 10Mb in size (wild guess), then it looks like 150 terabytes would be enough to store the entire web.
But I want to know more:
How deep does this archiving go? Are they going to store every single page and image of every single website?
How much storage space is required for the whole web? Wild guess: A recent
What software/OSes are they using for this project?
When do we get to see the archive?
If you say "I'll probably get modded down for this..." then I will mod you down.
You know, Google is doing something similar. They have copies of websites cached from the last time they crawled it.
On topic, though, doesn't this threaten to change accountability from people who post commitments on their sites that they cannot meet? In the past, these people could change their website at will, and since there wasn't a physical copy, there was no evidence of the previous comittment.
I anticipate this also being used in court. Think about it, if someone sues for libel, the evidence could be available in the 'snapshot' archive. This converts their project into a legal document, and means that the company doing this net archiving could be in danger of contempt hearings if they don't take extraordinary measures to ensure the integrity of the data.
I don't know, this sounds like an awfully big responsibility, and I hope this company has a good bunch of lawyers.