Google Code was never good either, but people gladly moved from SF to Google Code. And when something actually good came along (Github), Sourceforge simply vanished.

Sourceforge was never a hero. It was a horrible website, with horrible UX. It always had pretty terrible ads and for the past couple of years it has been repeatedly abusing its users' trust, serving malware, been involved in countless drama.

Glad it's gone.

Everything internet sucked around 2000. Search engines where either semi-curated listing or covered a fraction of what was a much smaller internet. Free hosting was a joke, affordable paid hosting was not much better. Your bandwidth at home was not good and anyway your computer was not the powerhouse it is today - you needed to upgrade just to run the next version of Excel, not to play the next AAA game. A lot of websites had terrible UX.

So Sourceforge was not great, but it offered a great package of functionality that basically solved in one shot all the problem of managing a OSS project and gave one roof where to search for new tools and libraries.

What happened is that they stagnated and their utility eroded over the time starting by Google removing any reason to have everything in a central place.

I disagree, and in fact preferred the internet of 2000 to the internet of 2015.

Google search worked fine in 2000: although SEO existed in 2000, it was much less refined and extensive than in 2015. The main problem in search results today is that profit-motivated content crowds out and makes it difficult to find not-profit-motivated superior content.

Not all the changes are bad, of course. Video in particular (not just bandwidth, but the convenience for consumers and amateur producers provided by Youtube) is much better (but text is more important to me than video). A computer with adequate performance is a lot cheaper in real dollars.

>A lot of websites had terrible UX.

I preferred the UX of the web of 2000 to the UX of the web of today. Searching a web page for a string, copying from a web page, then pasting, scrolling, use of keyboard shortcuts, bookmarking (some sites today have all content sharing the same URL, making bookmarking virtually impossible) all were simpler and more consistent across web pages, therefore requiring less deliberative thought to use, and IMHO thereby better.

In other words, on the web of today, I have to direct my attention to the mechanics of the web page, e.g., how to dismiss the interstitial asking me to follow them on Facebook or whatever so that I can start to see the actual content, which keyboard shortcuts have been overridden by the site, how to scroll and whether scrolling even works. In the web of 2000, I spent more of my time and attention on the actual content (particularly, textual content).

For those who remember the web of 2000: what about the UX of the web of today do you prefer? I have a theory that some people are more easily bored than others and that those people appreciate the web of today because the increase in variety, color and motion (of elements on the page) helps keep them from getting bored. Is that it?

(I remember the Sourceforge of 2000, but have nothing to say about it.)

ADDED. I neglected to consider that Wikipedia did not exist in 2000, which significantly changes the balance. I still disagree with "Everything internet sucked around 2000". In particular, I miss web sites created by individuals not expecting to make money from their writings. Although many (most?) of those web sites still exist, they've become vastly harder to find with a search engine because of content farms and other professionally-produced professionally-SEOed content.

The majority of banking sites required you to use ActiveX, very frequently the shopping websites would use weirdo content plugins or have their content in PDF.

Those were the times of multi-frame abuse, webpage with sound and no control, site trying to be clever and disabling right-click. Pop-ups everywhere, site hijacking your browser window and resizing, moving it around. Stupid cursor that make your computer grind to a halt. Site designed for 640x480 and nothing else.

Remember the "blog"-like personal website on geocities and the like ? Remember the portals ?

What I remember most from the time is how much superior I thought the newsgroups were to find anything.

A lot of these still happen, except they are using different technologies and look slightly different. We have parallax scrolling, Bootstrap-like modal windows, unlinkable single-page apps, websites that screw up "back" button and above all we have pages that require you to download several megabytes of junk and execute tons of JavaScript just to read several kilobytes of text.

I've tried eg Yandex and it's not much better, mostly just different (although they do return a good amount of results for things that Google's thought police have vanished). The unfortunate truth is that with the "internet" entering pop culture, pop culture has watered down the Internet.

I was thinking it would be interesting experiment to take adblock one step further and hack something up that retrieved all the Google results and filtered out everything with any advertising. It seems like I'd come across those hard-wrought dense-knowledge textual pages far quicker, since the author's motivation is to share knowledge.

Any time you have a "theory" about other people that implies that you have a natural superiority to them based on your belief that you have superior taste, you're almost certainly full of yourself

What part of that or my other comment implies that I think I have superior taste?

Also, in truth, I'm just tired of the "everything new is shitty, everything old is grand" comments on nerd sites, so I call them out habitually.

They've been in trouble, effectively since VA Linux's stock crashed...

If we'll look down in history, I think there are very few villains that have acted as heroes when starting out. The signs that such companies are assholes are in general there, from the start. As an example, I don't remember a time in which Sourceforge was the hero, as much as I try. Since as long as I remember Sourceforge, it was always the "necessary evil", the nuisance you have to deal with because bandwidth was (and still is) very expensive.

No offense to you personally, but this is a case where it's about grey hair more than ability.

There was certainly a time when Sourceforge was the wham-bam-snickety-snack boom-blam bomb. It was when Slashdot was the Hacker News, when "LAMP" was a new term, when 10Gbps across the U.S. (vs. your living room) was a Big Deal.

> If we'll look down in history, I think there are very few villains that have acted as heroes when starting out.

This is called playing "I told you so." In reality every organization begins with high ideals then adjusts to reality. The elegance of this transition defines how villainous they've become. Consider McDonalds and Whole Foods. Both seemed delightful at the beginning. We see how McDs is now cancerous (or at least diabetesacious) and Whole Foods is anti-union, but surviving anyway, a "mundane evil."

You can't predict which company will become evil or not.

But the underlying point is similar to "you are the product". People chose Sourceforge because it offered free hosting. It was for a long time the "default" choice, back in the early 2000s. They had banner ads, usually from other Linux companies, and that seemed to pay for it. It was easier to use than FTP sites.

The question always remains: who is going to pay for it? If a company is sinking, then they have only two options, to close it down or to engage in ever-shadier chasing of monetisation options. Can you imagine the outrage if they'd put up a paywall? It's basically the same politics as has made "free to play (with gouging IAPs)" the default mode of paying for mobile apps.

I do hate the "free to play (with gouging IAPs)" model and I actually prefer ads with an option to go ad-free. However even with this business model there are heroes. See for example Path of Exile, a free game in which the in-app purchases do not have an impact on your ability to win: https://www.pathofexile.com/

I do agree with you, the chosen business model makes the difference. I do believe that ads aren't inherently evil, it's just the status quo that is. Yes, today ad-supported means that we are the product, but this product has a free will, a fact that some companies are forgetting. At the very least there are some signs of change. Google for example is at least trying to play it both ways - you can be a paying customer of Google Apps and they'll introduce subscriptions to YouTube for going ads free (with the ability to cache content for offline viewing on mobile devices being the incentive needed for regular folks).

If Github ever pivots to selling t-shirts, run.

Ultimately someone needs to pay some bills though, a few transactions later and this seems remarkably far from their start. It's embarrassing and shocking.

The logical implication in that line flows in the other direction. It's not that every villain started out a hero, but that every hero (that survives long enough) ends up a villain. Of course, it's untestable as there's no limit on "long enough", but then again, it's just a line from a movie...

I remember a time when it was a fantastic resource, that it was one of the primary places to get FOSS - my earliest bookmarked link is for AFPL Ghostscript from 2004. IIRC I started using it around 1999 when I first installed Linux. There was Freshmeat too but that IME had far fewer of the major projects. Searching SourceForge was the best way to find projects for a particular need at the time.

The last FOSS project I downloaded there that isn't available elsewhere was last month.