I’ve my own grudges as well. I do disagree when Paul said “Y! doesn’t have smart engineers and hacker culture” but i’ve issues with the way projects are being executed there. Every now and then you are asked to change your projects. I’ve no issues if my current project is complete or shutdown but pulling someone out just because they like to be “agile” is definitely beyond my understanding. I’d to depart when i changed more than 5 managers and 3 teams within a year. Sure, i still miss a lot of other things about Y! including the great people i got to interact with. Not sure if i was just one off case, but i heard about this “agile” philosophy from few other co-workers as well when i was there.

Be aware that Yahoo bought Inktomi after the period pg was talking about. That is what turned into search, and part of the reason why Yahoo bought them was to import much needed technical talent. (That they later tried to spread through the company.) Therefore the culture in that group is not likely indicative of what pg encountered.

PG has his own opinions. Some of them are true, and some are not. I would encourage people to open their eyes that PG is not a god. Instead of agreeing on what ever he wrote in his essays (and accept them like the holy grail), think for once that it may be crap, like other other people’s essays do.

I worked at Yahoo! and later Google. I ended up in Yahoo! via the FAST → Overture/Altavista → Yahoo! route. Nobody remembers this now, but FAST, back in its day, built a web search engine which had a bigger index than Google and which was on-par with Google with regard to relevance for a brief period of about 6 months. I know a thing or two about working in a hacker culture.

The main problem with Yahoo! was its top management. It didn’t have any. Or let me rephrase that: Yahoo! never had management that understood the importance of technology, they did not have a realistic view of the quality of their technology and products and they had absolutely no idea how you cultivate a excellence.

You may think that Yahoo! had a hacker culture, but in comparison to FAST and Google, it did not.

In my very first meeting with Yahoo! we met a VP who summarized Yahoo! as “we copy what others do and then greatly improve on it”. When I give talks and mention Yahoo! I have a slide that illustrates what I thought of Yahoo after that meeting. The slide has a picture of a man in suit-pants and a shirt with his head up his ass. You might have seen this picture on the web.

I was lucky enough to work on a very good team with some brilliant people. Unfortunately Yahoo! is not a company where brilliance is properly cared for, so rather than allowing centers of excellence to grow organically by hiring top notch people whenever you can get them, Yahoo! imposed arbitrary headcount restrictions. (Finding top notch people is harder than you think. Believe me).

they apparently still have the idiotic bureaucratic headcount restrictions. which might have made sense if everyone on the team was mediocre, but makes absolutely no sense when you have a exceptionally good team that could do a lot more for the entire company’s bottom line if they were only given the proper resources.

The team I was on is easily one of the most important and productive teams in yahoo! and remains so to this day. But yahoo certainly has not done them any favors. they have not grown significantly even though they should have been allowed to, and things are very static. There is no real opportunity for a career — unless someone leaves and you manage to grab their spot. And I am not merely talking about promotions — I am also talking about lateral movement which is vital for keeping people fresh and inspired.

I left Yahoo! for Google (and I was followed by half a dozen other people from the same Yahoo! office). Not so much because I disliked Yahoo! (in fact, I loved working with many of the people there), but because there really was no future for me at Yahoo. Or anyone else for that matter. I had been part of the same team for 6 years and it was pretty obvious that the situation was completely static. The same people in the same roles with only minor adjustments, zero growth potential, and a top management that treated us the same way you might treat a bunch of factory workers.

I’m still proud to have worked on the team I worked on. They are good people and I would love to work with many of them again if the opportunity arises. But the ineptitude of the upper management, from the executive level and a few levels down, is heartbreaking. They are actively pissing away opportunities by consistently hiring clowns for middle management as well as for key executive roles.

Most of what Graham said resonated with my own experience.

I was down in Sunnyvale at Yahoo headquarters to give a presentation on Drupal to a group from the Linux User’s group there. I met an Indian kid, a yahoo programmer, who was really excited to be working there, and he had facilitated the meetup with pizza and free drinks paid for by Yahoo, along with the space given to us by them. His manager was pretty encouraging too and even mentioned that they were hiring on good programming talent. I got a ride back from him and another facilitator to the CalTrain, and they were talking about how the culture was changing a lot and there were some good projects being hacked on that they were really excied about.

So, I think I might agree with some of the opinions mentioned here, that Yahoo might finally be getting it and shifting gears from when pg was there, but it would be great if some more Yahoo programmers could share their opinions.

This question is never going to be resolved, because Yahoo is so big and sprawling and has had such a long history. What is IBM? The army of salesman in identical suits singing songs about the company? The people who had a lock on all business software for decades? The ones who really made Java a force in the industry? The brilliant eccentrics they have working on applying contemporary physics?

You are lucky to have experienced what were, in my opinion, two out of the three functioning areas of Yahoo, which are/were: Yahoo Search, YUI, and Flickr. Search works because of the imported culture from Inktomi and a recognition from the very top that this had to be excellent. Similarly, YUI is small and focused and they made brilliant hires at a time when even Google wasn’t taking Javascript seriously. Flickr doesn’t have quite the technical chops as both those teams, but has a fantastic focus on product and unique methods of iterating quickly.

My experience at Yahoo was very much like Paul Graham’s. One of Yahoo’s problems is that due to their success, money keeps raining down on executives and people keep getting paid, no matter how incompetent. Technically, most of Yahoo is lost in the 90s, batch-processing things with Perl scripts and human-generated keywords. Meanwhile Google has created entire platforms to automate infrastructure at the datacenter level and relies machine learning algorithms. This strategy is now available in bits and pieces that you can cobble together on the open interent. So while Amazon can spin you up a hundred servers in an eyeblink, most Yahoo properties still have to ask David Filo (yes, that David Filo) personally for more servers. One gets the feeling that Filo regrets that he isn’t personally racking servers any more.

The ideal Yahoo property is something like omg.yahoo.com, which is a repackaging of celebrity news, design and product-focused, with relatively boring technology powering the whole thing. It’s very profitable. Yahoo would rather have a hundred of those than one Flickr.

Nevertheless, Yahoo executives often make the mistake of acquiring hot startups, thinking they will infuse the company with talent and new ideas. As far as I can tell this has never worked. Yahoo Search, YUI, and Flickr remained apart from the rest of Yahoo, controlling their own hiring and technology. Everyone else was integrated into Yahoo and was shredded to bits by the culture. It only takes one or two pig-headed moves by executives and the real talent starts leaving. Or, like Delicious, you can be scuttled by the imported Yahoo engineers who throw out everything that was good about the original and try to remake it according to their incompetent practices. In some cases this has been deliberate, as when executives starve an acquired startup of resources, hoping the founders will leave, so that they can take over.

For most of Yahoo, the order of importance goes like this: executive whim, designer caprice, advertiser needs, and technology ranks a distant fourth. Rebel elements within Yahoo tried to institute Hack Day as a way of showing how technology could lead in its own right, but management has so far treated this as an exercise in employee morale, no different than a talent show or employee picnic. The organizers of Hack Day have tried for a long time to get the projects that win some funding, but executives have laughed — literally laughed — at the idea.

Agreed things are not so black and white as pg states, but I do observe a degrading ownership of engineering in the actual product. Though there are ofcourse exception, average engineer Joe is more worried about his code than his user. [more thoughts – http://bit.ly/9lctmC]

While there were isolated teams that did well within Yahoo! I don’t think I would count the old Inktomi team → Yahoo Search among them. Yahoo! Search suffered under a string of inept leaders from the CTO level down during the critical period between 2002 to 2005, and for all practical purposes it was definitively game over in 2005. The only glimmer of light was when Yahoo! acquired Overture and got FAST in the bargain. They then had 3 search engine teams: Inktomi, Altavista and FAST. The two first were vastly inferior, but due to politics it was decided that search was so important that geography was more important than competence. So they chose Inktomi with a mishmash of AltaVista sprinkled in for good measure. Had the management been competent, they would have invested heavily in the FAST team and let them run the show. Instead they ended up with nothing and eventually most of the brilliant people left for Google and Microsoft.