Historical programming-language groups disappearing from Google
lwn.net
lwn.net
But what happens when they get bored with map data and get rid of it?
He had been ordered to turn over all of their historical arial archives for scanning by Google, and then told the USGS would no longer do arial scanning since Google was doing it. But there was no agreement for Google to turn over their arial scans back to the USGS.
At the time we all told him not to worry, Google would never remove data it had collected. Looks like he was a lot smarter than us.
Googlewashing - to proclaim “Google would never ...”
It's not just celebrities, so many independent artists are putting up their talent on Instagram and I don't have access to any of it because I need an Instagram account for that. Instagram web version is forcing to sign up if you scroll 1 page down on a profile.
Sometimes I feel like we need to build cutting edge decentralized applications that will burn these walled gardens to the ground. /rant
The only thing that has improved is how easily a celebrity can reach millions
Celebrity is a reference to the fame and wide public recognition of an individual or a group
Check out https://bibliogram.art
Yes, please! More and faster.
Centralized services are easy to build, because they offer an obvious location to do some of the things that are tricky to do in a distributed fashion. They are also, by the way, far easier to for small numbers of coordinating people to control, which makes them popular with corporations, authoritarians and sociopaths.
Decentralized services will rarely be the New Shiny that attracts all the 14 year olds for a few minutes. But, unlike email, you never hear anyone whining that Myspace won't go away.
The reason central platforms win is because they have to be dead simple to use in order to attract any users. Decentralized platforms get their initial users because of how cool the technology is, but those people (people like me and you) aren’t UX experts and don’t prioritize UX.
It has to be easier than the central platform, and the central platform has the benefit of millions/billions of dollars to throw at it. Which means the decentralized platforms have to work even harder to overcome that. It’s not impossible, but it does require engineers to overcome their desire to build cool things and instead focusing on building a user experience that’s better than what Facebook/Twitter/etc can provide.
It seems like the Mastodon developers look at email and think “if it works for email it’ll work here” and don’t understand that people deal with email because they have to, not because they want to. I don’t want to have to change my email address when I switch providers and I don’t want to have to move all my stuff if my favorite Mastodon server decides to shut down.
That’s not a solution that’s just another problem. It’s bad user experience.
Not sure how that's worse than something like Facebook, where you literally don't get that option. If you want your asserted identity to be reasonably secure and easy to assess for other users, you have to find a trusted host or do your own hosting; that's no different from any other service.
Seriously this is exactly what I’m talking about. This is a textbook example of what I’m talking about.
No end user hosts their own email service.
I ran my own for a while using mail-in-a-box but ended up moving to fastmail because I didn't trust myself to maintain the setup. I need my email system to just work and that's likely the same for the majority of people.
(I validated that lack of trust in myself even with fastmail by not realizing for almost 3 days that I let my domain expire, thus causing emails to bounce with no way of me knowing that was happening)
This isn't difficult, per se, but it's not as easy as signing up for Twitter for the simple dumb reason that you don't need to make that choice. And the concerns about "wrong choices" aren't entirely unfounded; the first Mastodon instance I signed up for was effectively abandoned by its sysadmin. The decentralization still makes it hard to search for new users to follow compared to Twitter (checking right this moment, when I click on someone I follow to see who they follow, it only shows me people they follow on the same instance); depending on the Mastodon client I'm using, it can actually be a little hard to follow someone even when I find them if they're not on the same instance I am. Again, technically none of this is super difficult, but for a user who isn't philosophically committed to the fediverse, tiny little frustrations start to add up quickly.
Most public mastodon instances usually have an about page that describes what their intended audience is. You can also look through the public timeline to see what users are saying first before signing up. If you aren't sure, pick a larger general instance.
Despite all that, in my opinion you'll probably get more mileage out of joining an instance run by someone you know and trust.
>What if I choose wrong? What if I need to move?
On newer versions of mastodon there is already a migration option to import/export your data between servers. https://blog.joinmastodon.org/2019/06/how-to-migrate-from-on...
Regarding your second paragraph: If you'd like to fix bugs in your chosen mastodon client, I'm sure that would be welcomed.
What if you don't know or trust anyone that runs a Mastodon instance? And don't have the time/means/expertise/motivation to run one yourself?
People really, really want to argue that decentralization of social networks doesn't make things harder, but eventually the defense always shifts to "well, you have to be willing to jump through a few hoops if you believe in the advantages of a decentralized/indie internet, which you totally should," because the truth is that the decentralized way does make things harder. Personally, I do believe in the advantages of the IndieWeb, and I do think it's worth jumping through those hoops. I just think we need to acknowledge those hoops exist, and always be thinking about ways we can reduce the friction for people who say "I like all those ideas in theory, but in practice it's too frustrating."
It is as if we are all becoming lazy and/or many of us don't realize the harm in giving all our info to half a dozen super mega corps. Most of these mega corps aren't even distributed in the world, they are all American (except tiktok) which is another interesting angle.
This is going to happen (already happening?) in the webapps/apps world too. There are so many no-code tools popping up - most will die, the rest will get acquired by the mega corps. Made a great webapp that is successful? Now you are stuck with bubble/airtable/shopify/whatever. I cannot name many no-code tools that lets you export your application to be hosted independently.
I feel like we are on a path where in a few decades, a dozen or two corporations will control every single aspect of our lives - online especially, and probably offline too.
The modern alternative to Usenet is private Facebook groups that never get indexed.
This is a matter of demand, not capability. It seems most celebs just don't care about setting up their own stuff, and, really, why would they? There are free platforms out there that give them huge amounts of reach. Most of these people just don't need their own website. They may come to regret that decision later, but it's their decision to make.
If a normal guy/gal wants to set up their own domain and website, it's not hard for them to do so, certainly no harder than it was in the 90s/00s, and probably a lot easier. The "no-code" stuff certainly has lock-in disadvantages, but you can simply choose not to use them if you want. Yes, it's more work, but it was always more work to do it yourself, and always will be.
It's obvious that things like Twitter and Instagram provide value to celebrities and people who follow them. It's just that there are some serious externalities not factored in.
Plenty of rights would involve in it (copyright, privacy, and so on). Also, all kind of crimes are another issue. It is hard for people to keep monitoring if contents are safe or not.
I think we need a network version of the War of Independence.
Where's the non-proprietory decentralized platform that lets me reach as many people as I can on Facebook? There isn't one.
Why aren't the social functionality of identity / friends / followers / newsfeed / etc. built into browsers in a standardized way?
Facebook is 16 years old. That was a lot of time to figure out an alternative solution, but all we have are experimental projects that rely on adoption that they don't have to be useful.
Corporations aren't going to change how they behave, but it's annoying that us techies are apparently incapable of beating them at our own game.
Newsfeed is RSS/Atom.
Identity / friends / followers are really one package, and it isn't a thing browsers can solve on their own, because people want the ability to do password resets etc. Also, decentralized identity is somewhat the opposite of this anyway -- people don't want to use the same "identity" for their parents and their friends and their boss.
The best way to do this is for sites to use email as identity, because it's common and gives you password resets, but people can create more than one and separate them as they like.
Which the technology to do already exists, but Facebook and Google made it easy and the free software equivalent takes several hours to get running. Which we could fix, but haven't (yet).
RSS is sadly not enough on its own without the other puzzle pieces. Private feeds are not really a thing, it doesn't let you comment on or like or share the article to your friends, etc.
ActivityPub solves most RSS limitations.
Because these compete with the interests of browser vendors, interests which finance a degree of development that dominates and ultimately stifles independent efforts.
Remember that Google pitched Google+ as an "identity service". They're now accomplishing this through Android, Doubleclick, GA, Gmail, and ReCaptcha, far more effectively. And sell ads on it.
Facebook isn't going to pay for social integration development by Mozilla: Zuck wants that pie to himself.
Channel monopolies would prefer RSS died and browsers (or apps) served their specific feed directly and exclusively.
I like trains, and I started a website back in 2001 for people to share their photos. It was reasonably popular. One of my drivers was taxonomy and archiving of images for future enthusiasts.
Today, it's dead. People post their photos on Facebook groups. They get attention, likes - all the stuff that matters to a human. A week later the photos are lost in the group, hard for anyone to find, no indexing, no exposure. The comments - from people who worked on the railways, knew people involved - useful to historians of the future, are fantastic. But if you can't find them, what point?
I get why Facebooks succeeded. For my site, I was a total geek: why would I dirty the site with anything social? Well, look who's laughing now.
By contrast, Facebook is at best like a magazine, at worst a radio phone-in about trains.
Reference works in the form of websites have amazing value in and of themselves. I don’t think they need to be measured by social eyeballs when they attain an outright high level of quality.
I happen to be particularly fond of a reference website that is a taxonomy and history of British traffic lights:
Not even a week if you consider the target audience for what's posted as opposed to the poster. Algorithmic sorting and infinity scrolling have pretty much eliminated the ability to go back and look at something you saw a few days ago (unless the algorithm decides to boost it back into your feed).
But... many of the same companies will fill your search results or fill affiliate pages with quackery ads just fine...
Someone else mentioned that you can't reach as many people from your silo-ed website as you can if you go through social networks. I found one way you could get best of both worlds - through Medium's import feature[1]. But I don't yet know how effective that is.
Here's a short write-up in case anyone's interested: [2]
[1]: https://help.medium.com/hc/en-us/articles/214550207-Import-a...
[2]: https://ketanvijayvargiya.com/58-setup-blog-and-email-on-cus...
They were sharing information with the whole world, but in an ephemeral medium.
The web, and internet, is not an inhospitable place for anyone without corporate backing. You can host a somewhat reliable service on a raspberry pi over your home internet connection.
In my country, all physical books and magazines which are published must be submitted to the government in X copies. The government then keeps an archive.
With webpages, the problem of obtaining X copies never existed. Why couldn't the government have archived webpages like it always did with books?
Services provided by a company typically don't survive the end of the contract to provide that service. If the company itself goes bankrupt, all services cease to be provdided immediately.
Typically the only organisations which can credibly commit to providing a service for more than a few years/decades is the government of a country, a well-funded foundation with a clearly specified mission, or similar.
However, if they create a monopoly on that data they have an obligation to preserve it, especially in the case of a corporation outright aquiring data instead of simply "out competing" for data. And as everyone mentions, of course they are in no way legally obligated to do so, but they are by any reasonable standard ethically obligated.
I do think that the government could and should archive data, but there is currently no system in place for doing so and likely will not be for a long time, if ever. Corporations would simply have to maintain the data that they already have.
Some of that is because search engines have simply stopped returning them in results even though they're still online.
They are still figuring out what a good way for archival of those is and are quite selective in choice what they archive, but they plan to expand on that
German page: https://www.dnb.de/DE/Sammlungen/DigitaleSammlungen/dgitaleS... English page: https://www.dnb.de/EN/Sammlungen/DigitaleSammlungen/dgitaleS...
Jeez, that's horrifying. Literally just giving public assets to private corporations.
If the effort to USGS could be quantified in a cost, I'd expect Google to pay USGS to make the public data available?
It does sound awful. I don't know what the right answer is.
But they can still fuck you over, like when they banned my GMail account for no reason, with no warning nor explanation.
1. A corporation is not a person. Corporations don't have rights, except inasmuch as the people within the corporation have rights.
2. The problem isn't that Google has access to the data, it's that USGS and the rest of the world no longer have access to the data, except on Google's terms.
The supreme court was wrong on racism, and it's wrong on corporate personhood.
If any entity with a plausible use case could and still can get that data at the cost of the copy, I don't see why not. The whole "copying does not deprive the original owner" meme applies particularly to such public assets.
Can you point me to where I can download this data for the cost of a copy? Didn't think so.
Google presumably also didn't have a website with a download button.
Yes, money and access are important. Yes, ownership has money and access implications. But ownership in itself is fundamental, and the money and access problems usually don't immediately follow the ownership problems, because if they did, nobody would give up ownership in the first place.
I've said it before and I'll it say again: corporations don't have the right to "innocent until proven guilty". We don't have to wait for corporations to do something wrong to do something about it. A lot of people on Hacker News seem to have this idea that we should wait to regulate until businesses follow incentives to the point of doing great harm, and then once they're when they do great harm, we should just say, "Oh it's not their fault, they were just following incentives."
I don't accept this. It's obvious that taking publicly owned data and making it privately owned will lead to Google placing that data behind a paywall, an ad-wall, or a simple loss of access if Google feels they can't monetize access. Even if Google maintains a relationship with the USGS such that we always have access, there's no reason for the USGS to spend our tax dollars to pay rent to Google as a middleman. We don't have to wait for Google to follow incentives to that point--we can see where this is going and we don't have to pretend we don't.
So, if Google has put their copy of the data behind a wall, or deleted it, it's no less accessible than if Google had never been allowed to make that copy.
Furthermore, I'd be shocked if Google just kept the original copies that the USGS gave them.
But we're just not smart enough to understand that, never mind make it happen.
Instead we prefer to cling to the bizarre delusion that billions of individuals with competing interests will somehow spontaneously self-organise into the best of all possible worlds.
However, to be fair, it has always been the most greedy and self-interested, with already the most disproportionate power to rig the game in their favor, that have been most vocally advocating this system. No surprise there, of course.
What fascinates me is how a majority of people, who certainly do not personally benefit from that system, have been made to believe that they do. Sure, political corruption, cultural indoctrination/propaganda, horrendous general education, and I can think of a few more .. but still I've always been amazed, how it appears to have canceled even basic logic reasoning among so many.
Who knows, maybe one day, it will turn out to not just "correlate" with an addiction to carbs/sugars, of which the country has plenty of problems with too. Junkies have always been easy to manipulate.
Until then, at least it still gives some hope that a growing number of people now realize that this system just doesn't work as it is advertised.
https://www.usgs.gov/core-science-systems/ngp/3dep/3dep-data...
Making digital data publicly available is pretty new for USGS. Just a few years ago archived aerial imagery had to be ordered by mail and it was a pretty lengthy process. Topo maps (the earlier equivalent of the DEM data to which you refer) were generally ordered on paper as well up to five or so years ago, but they're in a lot more popular use so more third parties got into the business of distributing them. I've relied moderately heavily on both for some of my research and was a very painful process until just recently to get anything older than current. In the meantime, yes, Google had it all at some point, but mostly stopped using it or providing it because they obtained better quality imagery.
Fortunately USGS now has a slippy map for topo and an admittedly rather clunky ESRI query service for aerials.
Keep in mind that when we talk about the EROS archives we're talking about data that goes back to the 1930s and earlier for some product types.
For a long time I got the topo maps from the website of a state government bureau that had conveniently run them through their own large-format scanner and posted the TIFFs - USGS didn't get around to it for years after. It's hard to blame them too much as they had a shoestring budget.
Actually, for amusement value, that state agency appears to have removed the TIFFs from their website and now says that you can order the topo maps by mail for $8 a piece, which is what I used to have to do. I wonder if USGS got mad at them, which is a bit ironic since they don't mention that USGS themselves only recently started offering them online for free. For additional amusement value EarthExplorer, the fairly new service that lets you retrieve aerials online, has a banner up that downloads are intermittently broken and indeed I can't get it to work at the moment.
That was poor negotiation by USGS Solicitor's Office. Libraries participating in google digitization programs negotiated to keep copies of their scanned materials in the Hathi Trust Digital Library https://www.hathitrust.org
Looks like he was a lot smarter than us.
If you would've asked me back when Google was new, and we all believed in "Don't be Evil," I would never have thought that Big Tech would end up being the Ministry of Truth and The Memory Hole.
Anything that is (no longer) of commercial value will be "phased out" and dismantled/destroyed. One might still stretch it a bit, by arguing that the commercial value of something can include its future potential value. But I personally know not a single commercial companies that ever choose that over short term cost reductions and "profit optimizations".
Luckily, there are governments who acknowledge this shortcoming and build structures to compensate for it. But when governments decide to leave (almost) everything to commercial markets, then the importance of anything and everything can and will only be measured by it's commercial (contemporary) value/profitability.
People have every right to vote for and support such a system. But then don't complain, when all that you will get is only what such system supports/provides.
A font of knowledge
I am a holdout.
(Not suggesting I am "smarter" than Go users, but I can forsee issues with Go being controlled by Google.)
Anecdotally, a close relative (and many others in her institute) designed entire curricula of learning modules for a government-owned nationwide technical college, back when online learning was newish, ~20 years ago (I think back when SCORM was fresh). These were tightly integrated into the traditional in-class offerings. A couple of years later a "trim the fat" government slashed internal capabilities and outsourced all "IT" hosting, management, etc.
All of the online learning modules (which would have cost millions in man-hours to develop) were literally handed over as "content" to a company who to this day offers them back to her institute under per-student licenses (that far exceed any "hosting" costs of these basically static resources) over a decade later. This company also profits off licensing to an array of pop-up online "institutes" that don't even approach the pedagogical context needed to ensure quality education outcomes from these resources.
Like a comedy of errors, from time to time some lecturer at her college will want to ask some question about the materials, their boss directs them to the company support (which is a paid service), after the issue escalates through the support tiers and they realise they need the expert knowledge of the author she'll get an email with the question, a process that can take days or weeks when the lecturer could have walked into the office next door and asked her directly, if the company hadn't stripped all author credits from the materials.
If the company decides to shift business models, or goes out of business, or is acquired and scuttled, these assets get blown to the winds.
There's a lot I could say about this situation, but essentially governments in general seem to devalue their assets at taxpayer expense, the IP of these assets could have been better handled rather than just giving it directly to the first company to win the contract all those years ago.
It sounds awful that Google has the best mapping data in the US. In the UK Google's data is awful, worse than OpenStreetMap and much worse than Ordnance Survey, the national mapping agency.
It's fine when you travel by car, but when I'm hiking through the hills I'm just walking through an empty square on Google Maps. Volunteer-driven OpenStreetMap is MUCH better. And there the data is actually open and safeguarded.
Governments should support that kind of project instead of corporate privacy-invading playtoys like Google Maps.
Blocking posting access to these newsgroups from GG is generally a good thing for those newsgroups.
Not being able to search the archive is the unfortunate collateral damage though. Google is not obliged to provide a Usenet archive, I suppose.
Formerly obtained deep links to the content also do not work!
If you formely cited a comp.lang.lisp article by giving a direct link into Google Groups, people navigating it now get a permission error.
NNTP is a wonderful protocol, arguably the simplest of the 4 mailnews protocols (IMAP, POP, SMTP, and NNTP). While it seems to share the same basic format as RFC822 messages, it actually tends to avoid some of the more inane issues with the RFC822 formatting (generally prohibiting comments and whitespace folding).
Unfortunately, the internet by the early 2000s started turning more and more into an HTTP(S)-only zone. Usenet itself hemorrhaged its population base, especially as ISPs shut down their instances (e.g., because someone found one child porn instance somewhere in alt.binaries.*).
Ultimately I think the web would have eaten Usenet anyways, but it's a shame; we were Freenix-competitive (I think I independently invented the INN history cache), and that was some of the most fun I've had doing systems engineering work.
I didn't start posting to usenet before 1999 and was a regular poster in a few groups from that time up till around 2014. Excluding spam, what was the activity level of groups before and after eternal september?
> It was extraordinarily difficult to keep reliable full-feed binaries
I don't understand why ISPs wouldn't just limit their newsfeed to the text only newsgroups? Did peering arrangments require one to also provide binary newsgroup access? IME, ISP and university news servers had mediocre binary completion rates at best. If someone wanted binaries, they could always subscribe to one of the paid newsfeeds that provided better completion. So I don't really see the incentive for providing binary access at all at an ISP/educational institution level.
What really took down usenet was when Andrew Cuomo, back when he was the state attorney general of New York, made a deal with several major ISPs to restrict access to child porn via usenet.
This lead to many ISPs discontinuing their usenet service, which in turn lead decreased the number of people posting to text groups. Within a few years of that happening, practically all the regular posters in the groups I used to frequent just stopped posting. Those same groups now only have spam posted every several weeks based on what I've seen via google groups. Prior to that, these groups had plenty of active discussions going back to the mid '90s and earlier.
The ISPs needed the binaries though, because that was all they were used for. People read the text on their Uni systems, where they got free dialup, rather than chew up their ISP connection time quotas.
At dialup speeds, the only practical binaries one could download would be images. mp3 files were practically the upper limit of file size one could download. Beyond that, articles would expire off the server before they could be downloaded.
Without broadband and better completion rates, which commercial ISPs didn't really provide (especially the latter), customers probably wouldn't really try using their usenet feed for that purpose when they had other alternatives for binaries.
I just used my ISP's usenet access for text groups until they discontinued it.
Vladimir Panteleev did, though, write a web interface to NNTP:
which is also freely available:
Probably a bad idea, according to what we know now about echo chambers in social networking.
(Though mere long article retention is not necessarily the best archive interface, of course.)
Disclaimer: I'm not well-versed in the solutions in this space. Maybe there is some NNTP cacher out there that also has a web archive interface into it or whatever.
https://github.com/DigitalMars/ngArchiver
and the generated pages:
https://digitalmars.com/d/archives/digitalmars/D/index.html
When we were working on the history of the D programming language paper, this was an invaluable resource.
I wish forum.dlang.org had a quick way of browsing just the top list of the most commented posts.
There is also https://www.eternal-september.org/ which I used.
AOIE requires no authentication. The Eternal September server requires account registration via the web site; then you use an authenticated NNTP connection.
There are other servers out there.
These sites do not provide any archive.
It's because, at the time, you don't know what information is going to be important and what is just garbage. Documents that are apparently useless today could become fascinating tomorrow.
That or risk future archaeologists thinking COBOL was some God of the time and the natives built large metal obelisks in dedicated worship temples.
Never forget that we do not know the future.
There are still interesting things to be learned from ancient artifacts.
And if not, what makes comp.lang more like the pyramids than geocities?
Yes: https://www.archiveteam.org/index.php?title=GeoCities
Digital data is not exclusionary in physical space like condos. And even random myspace pages with hacked stylesheets show the common culture of an era.
Personally I'd like to be able to link to my own posts from that time, for when people asked me what I used to do. But I can't find them any more.
These groups are mostly not code. They are conversations, design discussions, ideological discussions, jokes, that sort of thing.
Like what we have now in social media, except back then there was pretty much only Usenet, and it had a very different feel than the current social networks.
They are where things ideas like the smiley, and free and open source software, and utopian ideas of internet culture were developed. All the early internet memes. And of course all the knowledge people shared.
Conducted in public at the time and thought to be archived for the long term.
You're right though that a decision will probably have to be made at some point about what to keep and what to toss (how big is YouTube, exactly? Are we really going to keep every video, in its original resolution, forever?), but this is just plaintext, it takes up almost no space. The decision doesn't even have to be made, since it's easy to find the means to store this, so why bother making it? Kicking the can down the road is actually the best decision in this case, since the people of the future will (hopefully) have a clearer understanding about what was important in our own past than we do currently.
Interestingly, when some people saved a great deal of the Usenet archives pre-Deja News, one of them said something to the effect of they wished they had prioritized saving social discussions and so forth because, by and large, saving discussions about a bug in a long ago version of SunOS probably wasn't very interesting.
Honestly even that sounds pretty fascinating:
It could help someone gather stats on the nature, frequency, and severity of bugs over time and across companies from another angle.
It could provide a fresh perspective on modern OSes by showing how historic OSes did things.
And it might be good material for a course on the history of software engineering practices, showing classes of bugs that have been eliminated, and styles of development and customer support that worked or didn't work.
Here's the article I was thinking of by the way. https://www.salon.com/2002/01/08/saving_usenet/
likewise many people are clinging to the local operating system rather than moving to the SAAS model.
so what happens if we lose the oldschool languages and platforms entirely, for whatever reason ?
if TBTF corporations are somehow hobbled or neutralized, we need old hand tools to build a tech newtopia from the rubble. if those tools are destroyed then we are beholden to a system that stands on very thin ice.
I second the need to rebuild from the rubble is often overlooked, especially by corporations driven by profit centered goals.
Those archives are full of useful and informative information.
Not everthing changes fast. Common Lisp has been around for 30 years basically unchanged. The discussions back there can be truly informative for today.
It does take time to wade thought it, but people have been collecting (via the google archive, when it existed, sigh) curated lists.
https://www.xach.com/naggum/articles/ https://www.xach.com/rpw3/articles/
1) Eventually, everything will be lost anyway. The original print of King Kong is gone. A fire at Universal Studios wiped out the masters for a lot of music at once https://en.wikipedia.org/wiki/2008_Universal_fire . Floods destroy family photos all the time. But those are examples of the forces of decay, of natural entropy, of error. The Library of Alexandria probably contained a lot of useless crap but also nuggets we’d want to know today. Information is memories, useful information is useful memories, and there’s no compelling REASON to lose it. Other sections of usenet history were wiped out when Google acquired it (a lot of comp.database.olap content I had a hand in) and groups of people just lost a knowledge base.
2) It’s not simply code that no one uses anymore. It’s a knowledge base on how and why, debates over constructs and usage that are useful beyond code-sharing snippets a la Stack Overflow.
3) There is an argument for letting some information get lost or at least super-obscure, but it’s hard to see this being a good example. Tide Pod Challenge videos come to mind. GDPR and right to be forgotten mandate something akin to information loss.
4) I posted this elsewhere but I’ll share here too: there was a comment made on the original article about preserving prior art for IP (patent) purposes. That alone is in the public interest. Irrelevant to your questions in general, but pertinent to each of them in this case.
Given the nature of Usenet, they were if anyone wanted them.
What Google is doing by refusing to publish the archive or even share it with parties like the Internet Archive is completely unjustifiable and anathema to everything they once stood for.
Couldn't a copyright claim (or something under the GDPR or UK's DPA) be used to regain access to those though?
Just because something is published to a public forum doesn't mean you relinquish your rights.
The DPA isn’t new - it was created in 1988 - and UK ISPs had Usenet/NNTP servers long after that.
They cared enough about to kill it.
If you look at the history, Google basically rescued the data from a collapsing Deja News, and made it available again. A nice gesture, which didn’t serve to benefit Google much in the long term.
If we want to preserve history then we can’t rely on for-profit companies. We need to instead fund non-profits whose specific charter is archival and preservation, like the Internet Archive.
(This was at the same time that there was a gold rush of IPO plays, hiring anyone who could spell "HTML", and plopping them down in slick office space, Aerons for everyone, and lavish launch parties, with tons of oblivious posturing and self-congratulating. But Google stood out as looking technically smart, at least I believed the "Don't Be Evil", since that was the OG culture, and it seemed a savvy reference to behaviors in industry and awareness of the power that it was clear they would probably have.)
That might be why it wasn't surprising to hear of things like someone entrusting a bunch of old university backup tapes to Google's stewardship.
This has played out with mixed results, and I think Google could be doing much better for humanity and for techie culture.
Brb archiving my Twitter posts
Well, you assume. Maybe it was just decentralized enough you haven't heard about it.
So I do think they have an obligation either a) to make the whole archive available for anyone or b) maintain it properly.
Properly means restoring the fast UI from around 2004.
Dejanews was the seed material for Google Groups, any profit derived from that (ads) was from content posted to Usenet by people who never intended for it to be used for that.
Sadly, even in 2020, nothing has yet replaced what Deja was at the time it got acquired and destroyed.
I'm curious what second or third order effects you think a usenet archive had on GG.
In addition to the search history, email content, geo location etc. G have for many people
It's probably not a good idea to depend on a public company to steward an important community.
Does the Internet Archive have copies of all the old stuff at least?
Which is sad, but expected.
That'd be an improvement.
Page & Brin retain controlling interest, despite their minority stake.
I - as most of us - have a personal google account, and our company uses a google business account. While I'm following news regarding google cancelling accounts at will, I fail to notice a reliable pattern: (alleged) fraud and other illegal stuff seems to comprise a good part of it, but at most 30-50%.
I treat all Google accounts as throwaways now and don't use the work email at all because I want to know that I can actually receive emails that are sent to me. That's a huge problem even without randomly losing access, because their spam filter has a ton of false positives and those emails don't get forwarded to my real address.
I tried to make a Google account for work use the other day, and got stuck at that point. Given Google's history it seems silly to use my personal account for work, or to connect the two accounts in any way.
https://c-command.com/spamsieve/help/turning-off-the-gmail-s
It is an awful hack. If they can’t be bothered to maintain their spam filters, they should at least let people opt out.
This is very interesting to me. I've used Gmail for 10 years now and I've found the spam filter to be nearly impeccable. I can't recall a single false positive. I can't even recall a single false negative, though I am moderately careful about who I provide my email to.
Now I'm left wondering if most people think about Gmail more like me or more like you...
I'm in their spam sin bin since a spammer managed to find an old test account on my SMTP server with a weak password and spammed the world for a day or so a few years ago. The problem is that I don't send enough emails to get out of the bin.
This isn't a problem google cares about, small senders with no reputation are basically screwed. I can deliver to gmail hosted accounts I've got a relationship with (personal & my own work address) but I can't reliably send to other email addresses at work.
a) Spy on people and sell the data to advertisers.
b) Use that data to directly push ads
That's basically incompatible with b2b services. Or consumer services. As a customer you're judged by how valuable the data they are collecting on you is. Which is less than a support call costs. That bleeds into every facet of their business. As such even if you pay them money you get the same treatment because they can't think any different.
Do they?
Income, gender, location, recent history of viewing specific types of pages, etc.
This is what I mean by limited form of access. Advertisers do not receive user information, but are granted the ability to use it when setting up ad campaigns.
Play Music has not been shut down (yet), and you can transfer everything to Youtube Music, which is available at the same price (and in my opinon a superior product).
Spotify is generally better than Play Music though, so it was for the best in the end.
Also, and this may be a bit of a tangential point, but the "deny the past because it has something bad" that Google has effectively done here is uncomfortably close to the set of recent and far more political events.
You just reminded me of a quote from an electronic music documentary 25 years ago. One of the Detroit techno artists insisted on taking the filmmakers to a historic theatre that had been left to crumble & turned into a car park:
"In America especially, nobody tends to care about these kinds of things. People in America tend to let this shit just die, let it go. No respect for the history. I, being a techno, electronic, high-tech futurist musician, I totally believe in the future! But as well, I believe in a historic and well kept past. I believe there are some things that are important. Now, maybe this is more important like this, because in this atmosphere, you can realize how much people don't care, how much they don't respect. And it can make you realize how much you should respect."
- Derrick May, DJ/Composer, Universal Techno (1996)
https://youtube.com/watch?v=tdox6H7FJBU&t=955s
The segment starts at 16:00 in the video and is about 2 minutes long.
You may be surprised that it's not just companies. It's not hard to find people who think it's better for old stuff to just be deleted.
In fact Usenet predates spam itself, since the first spam (Canter & Siegel) was on Usenet itself in 1994 (I was there).
I think you have to register. Not sure how much history is there.
I have no idea how useful the collection may prove to be. I found 'comp' but it doesn't offer a webpage view, just a link to download a file. https://archive.org/details/usenet-comp
https://groups.google.com/forum/#!forum/comp.lang.forth
> Banned Content Warning
> The group that you are attempting to view (comp.lang.forth) has been identified as containing spam, malware or other malicious content. Content in this group is now limited to view-only mode for those with access.
> Group owners can request an appeal after they have taken steps to clean up potentially offensive content in the forum. For more information about content policies on Google Groups, please see our Help Centre article on abuse and our Terms of Service.
There's no content available for me.
Looks like there has been (likely automated, nearly all of them are the same Italian phrase) mechanical legal complaints and it probably caused this instance of automated blocking going wild.
As an engineer I can understand the desire to automate everything, but please at least have some heuristics to detect this kind of easy-to-detect mechanical behavior before giving the model a full authority to block anyone it doesn't like.
A Genoese lawyer has been a victim of harassing and heavy doxing for some time, you can find many twitter accounts accusing him of paedophilia in cahoots with epstein, berlusconi, the pope and so on (no, I'm not kidding; clearly the stalker has obvious mental sanity problems).
The stalker is very prolific and is wallpapering the internet with his copy-paste-accouse in every corner, from newspaper comments to ancient forums to usenet. The lawyer report and ask for removation where he can but also he does not seem very worried because it seems that this issue goes on from two years ...
I don't think I can say the name of the subjects in question but in any case I'm archiving the harassment accounts before proceeding with the report, then I'll try to get in touch with the lawyer and see if he can request a new, less "coarse" censorship.
Was I naive in thinking that The Internet Archive would have long archived this type of thing?
If you want to look, you might start here: https://archive.org/details/usenet
Thanks for posting this, it reminded me to donate again to archive.org, which I just did.
I use ‘culture’ to include anything creative, anything that we experience as humans. Everything should be preserved, schools should be well funded, as should the arts.
> In 2009, Ron Garret published a 700MB archive file of all of comp.lang.lisp
Of course the downside of Usenet was most people expected conversations to disappear after a couple weeks or a month but there was always some jerk that kept everything and refused to delete anything.
They sucked the air out of advertising (in cooperation with Facebook) leaving none for others. But I consider that a small loss.
Microsoft did that for operating systems, productivity software, stalled the web with IE6, and more.
Google is capable of much more damage, for sure. But they haven’t done that damage just yet.
That is changing extremely fast.
I beg the difference, Gmail have not changed much since I signed up 16 year:ish etc
They are all the same, as soon as competition goes away, this happens.
For most services there's a better alternative (e.g. DuckDuckGo with bangs for search, native apps whenever possible, markdown for notes and texts) - in general this was a learning process for me to distrust systems that control me more than the other way around by creating a lock-in to an ecosystem.
I feel it's worth taking the time to understand systems a bit more. Once the learning is there, much of the convenience that Google offers can be replicated through good processes and automation.
If Gmail required emails themselves to be in a special format that broke other MUA and IE6 wouldn't render standards compliant emails in a way you could read. That would be analogous to what IE6 was up to.
Gmail is as notorious as IE6 was for its rather poor support of HTML and CSS in email.
(Disclosure: I work for Google, speaking only for myself)
Android, GSuite, Chrome
Though Android's marketshare today is extremely similar to Windows compared to macOS early in the period for which Microsoft is criticized. Similarly, Safari and Firefox combined have low enough marketshare (currently estimated around 21%) versus Chrome estimated 65% (and growing) to make real, very concerned comparisons to the early parts of the IE6 era, if not its peak (yet).
The difference is that it is still early and Google hasn't killed their competitors yet. I can't tell you whether or not they are "trying" to, just that I agree with the above poster that Google has the potential to be worse, and possibly history repeating itself on some of the exact same product lines.
Though the same people who complain about the loss of profitable journalism also run adblockers and sub/unsub from major newspapers over their op-eds.
It's a vicious cycle kicked off by Google and co.
You need a lot of principles and some money, getting those things together in the same place seems hard.
Was news in the past better ( more comprehensive? more verified?), or just better presented?
1. I'm somewhat nostalgic to physical newspapers for a couple of reasons, I think because they were physical and required a little bit more effort. They felt more authoritative and verifiable, that's the nostalgia. However I think there was another benefit to the journalists, because they were crafting a physical body of work, I probably naively believe it instilled a sense of duty in that work. Personally I feel like if my product (not code) was a fleeting piece of information, easily changeable and forgotten among a deluge of other work. I would not feel as obligated to honor the craft. So in short, less comprehensive and better verified and presented.
2. News was concentrated, I don't think this was necessarily a good thing. This is somewhat contradictory to 1. But I think this concentration was an easier to accept narrative and speaking to my above point, that presented a level of even-headedness to the whole affair of understanding the world and our place in it. The firehose that was unleashed with Google was good, but it then signaled the death of lots of rational measured thought and brought us to where we are today.
News in the more recent past tried to be a little more hands off of politics. The market dynamics were usually such that you wanted to sell both to the left and the right. Some newsmen idealized this and sold the product with the idea of an allegiance to fact, and the idea that the facts would speak for themselves. Opinions were indulged on the editorial page (where advertisers fear to tread) and they could be safely ignored. Opinions were avoided elsewhere. Sensationalism was for tabloids; major papers achieved respect and a reputation and premium prices with their restraint.
News of the present finds the market dynamics have reverted; overt partisanship and sensationalism drive reader engagement again — but we’ve got a ways to go before we have overtly political journalism everywhere and widespread cases of yellow journalism like in the past.
Easiest example is with RSS - entered the RSS Reader market for free and at a loss and effectively killed competition because you cannot compete with that. Then subsequently killed Google Reader. This chain of moves essentially drove RSS to being obsolete which in turn made everyone far more reliant on Google and social media.
Now extend this to other products that they’ve started for free and subsequently killed. It’s not the same as embrace, extend, extinguish, but the result is the same. You kill off competition and stunt progress.
It’s mostly that RSS isn’t monetizable as easily as web pages. I think FB and Twitter dropping their feeds had a more significant effect; regardless, RSS was always niche.
For me, the key is that Google Reader actually killed off competing products during its existence. Then when they killed of Google Reader, that stunted the ecosystem because anyone who wanted to provide an alternative was starting from the ground up. All the time that could have been spent driving RSS forward was instead spent on catchup.
FB and Twitter dropping their feeds may not have happened if the RSS ecosystem evolved to benefit them in some way.
By inventing XMLHTTPRequest?
To fix problems caused by Google, you need to change the principles of competition law. Microsoft was knowingly doing lots of stuff that violated laws. It was just very hard to prove it.
I mean, it was all in the news, trade magazines, business journals. Blackmailing OEM's, intentionally breaking things and making them incompatible. At least the legal battles are documented somewhere and Wikipedia has something about them, but they were just the tip of the iceberg.
https://en.wikipedia.org/wiki/Microsoft_litigation
https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor....
https://en.wikipedia.org/wiki/Browser_wars
There must be book somewhere.
Dan Gilmour's articles in San Jose Mercury news from 90's should be somewhere.
Basically small software startups had to have Microsoft Strategy. They had to find way to stay out of Microsoft radar or MS would steal their work, their developers or block them. You sue them like Stack did and MS just stalls few years and pays few millions in damages. It was worth of losing in court to protect monopoly.
Big OEM's like Dell had to do what MS said or MS would up their price. It was straight blackmail from monopoly position.
These groups should be putting more effort into federalisation and decentralisation. Make it possible to store all of this data in a distributed fashion and stop relying on a central authority for archiving purposes.
The problem is that there is no other searchable archives.
I had a UUCP news feed from a local internet provider when I was in high school, back in 1993 or so.
Like when Google decided it's going to host comp.lang.c, can there be only one comp.lang.c on the internet, or can someone else start hosting comp.lang.c as well?
[ SELF FOOT SHOOT ] 1000 REP
Aside from the spam, it gradually switched from passionate but respectful debates to name calling and plain insults from newbies to what remained of the veterans.
One could read very long arguments between Elizabeth D. Rather, CEO at that time of Forth, Inc. which she founded with C. Moore somewhere in the 70ies, and Jeff Fox (RIP), who was working at that time with Moore; Moore left his first company to pursue its adventures in hardware, making different "Forth processors", which eventually led to the RTX2000 which powered, notably, the Rosetta probe.
This is really bad marketing
The support ticket was deleted, so I guess not.
Once you see things in this light, the new flavor of the month online service just doesn't hold any allure.
> Has anyone (EFF?) considered the aspect of destroying evidence of prior art in the public domain?
I think there’s a case to be made for stewardship of these groups for that reason.
I guess there was this unjustified assumption that google only adds & never subtracts.
Not safe for work!!!
In google’s case you can see this boredom on Android, on the number of products announced and casually killed (they might be excellent standalone products but can’t move the needle on earnings for the benefit of Wall Street so why bother?).
Contrast this to early Intel in the Grove era: they were on top of the world with the memory business so they pivoted to something else. Google has had the same two products for almost 20 years. The later Intel has been more like that.
Another contrast: they don’t know what to do about the advertising downturn, so are cutting back on hiring and such, while FB is trying to double down.
1) Hiring standards have drifted downwards over the past 15 years. Google used to be super-elite, compact, do-no-evil, massive-profit-per-employee. It's now a 140,000 person organization, and at that scale, standards just aren't high. You have a team of dozens of incompetent people doing what one person used to do.
2) With COVID19, ad revenues have crashed. It's not clear the impact on Google.
3) The smart, ethical folks on top (folks like Larry, Sergei, and Eric) are gone, and replaced with professional managers. They were smart to pick an internal CEO, but most of their executive team comes from places like Microsoft, Oracle, or Morgan. Having known a number of professional executives, the key skill is climbing executive ladders and moving into positions of power, not running successful companies.
4) Their products are increasingly starting to crash-and-burn, especially in B2B. Their culture relies on automated systems over people, and their automated systems have taken down tons of mission-critical businesses. Automated works well at 1000 people supporting 7 billion in B2C (small elite team model), and not so well for a massive, 100k person company.
5) I've switched mostly to non-Google products because they're better for what I need. AOL was massive too at one point. Losing the tech edge is not good. I still use gmail.
On the other hand, their revenues have continued to rise exponentially since they started. So perhaps they're doing fine?
Time to de-Google the whole Web.