I think we can all recall the excitement when John made the initial announcement for what was then called Quake Zero. A free, browser based, updated version of Q3? What could be better! Much time has passed since, Quake Zero got renamed to Quake Live, closed beta came and went and we have Quake Live in as 'stable' a state as we can expect to see it for the foreseeable future of the project, minus minor tweaks and fixes.
As this happened, a company called S2 announced a similar venture: they will be creating an updated version of the other game I play, Dota, called Heroes of Newerth. HoN attempts to improve on pretty much all the same facets of dota as QL is attempting to improve on Q3: an easier and improved way of joining games, a cloud-based approach to replace the private server based approach of the previous generation, stats tracking, and matchmaking (although this has not been implemented in HoN to the same extent as in QuakeLive).
Despite the similar goals of the two projects, they do differ in a few important aspects: HoN is a complete standalone executable, compared to QuakeLive's browser based approach, HoN is not made by the original developers of either warcraft 3 or Dota, and HoN is not free. The largest difference between the two seems to be that HoN has received overall positive reviews from Dota players, and has already seen a large part of dota's massive playerbase migrate over willingly, whereas the Quake 3 playerbase has been more-or-less forced to migrate over, and has done so with the standard amount of quake-player whine.
I find it perplexing that the response to the two games have been so different when the end net result for both have been (at least to me, perceivably) the same. I have since begun to take a closer look at both games, to determine what the major factors were that seperated HoN's good reception from QL's mediocre one.
The first, and IMO largest difference between the two is HoN's standalone approach compared to Quake Live's browser based approach. I work as a developer myself, in a largely data-driven business environment, working with standalone legacy Delphi applications. About two years ago, we hired a 'web guru' who took a look at our system, made a quick demo in a week or two's time, showed it to our CEO and said that our entire legacy system can be replaced by a web-driven interface in 6 months to a year.
Naturally, our upper-level management was enthralled by this idea - the basic demo had been done in no time at all, the new interface would be slick, easily updatable, and easily accessible. To cut a long story short, two years have gone by. The web system has gone from being 90% to 92% complete the past year. Our 'web guru' himself decided that it was not a good idea in the first place, and is now campaigning for a .net-based standalone application approach to resolve the same problem as the web-based approach.
I think the same thing happened at id: someone managed to wrap quake3.exe in a browser plugin ("there are one or two small issues, but this is a brilliant proof of concept and I'm sure we'll be able to iron out those without any trouble"), showed it to the bosses who loved the idea, and id went the browser based route. However, developing and maintaining a complex web application is tiring, gruesome development work. The common 90/10 project paradox is exaggerated for web based work, and besides that it is horrible work that few like to do. I think id grossly underestimated the amount of work the web and data portion of the project would entail, and I believe this is responsible for the slow progress and sparse updates we have had as of yet.
Compare this to S2's approach: they have complete control over their interface. The initial bit was more work, I'm sure, but since then any change or minor tweak has been much easier, compared to having to hack every single change to work on what amounts to 3 different virtual machines for the web-based approach. The experience of using the interface is smoother as well. You can access the same stats views, buddy lists, irc-style chats and all the rest in game, in the same format as you would between games. All this adds to a very smooth game experience, rather than the slightly-disjointed QL experience. I think this was a mistake on iD's part, and one that they will correct if there ever is a Quake 5.
Another key difference is how faithful either game is to their predecessor: HoN, while being extremely similar gameplay wise, differs greatly in look and feel. I think the greater the difference between a game and the followup, the better. When a game is as similar as Quake Live is to Quake 3 (or CSS is to 1.6, or Q4 is to Q3...) the experienced fans tend to focus on the small things that are lacking from the update rather than the limited amount of new features. HoN has this 'wow, new game!' factor which keeps it interesting and exciting enough while the dev's iron over the small bits of trouble that Icefrog and Blizzard have had years to take care of. On the other hand, when you started up quakelive for the first time, you saw quake 3, except without brighskins,timeouts,locking teams, and all the other missing features. This has made the community focus on the negative alot more than the positive for Quake Live.
On the community front, S2 has been alot more open and communicative with the HoN community than iD has been with the Quake Live community. However, due to the nature of Dota and it's players it is easier to communicate with them. You can let a bunch of Dota forum regulars loose on /b/ and they will organize themselves into different sections, with mods and everything, within weeks. Quake forum regulars would have the opposite effect on an organized forum (cough esr cough). Still, there is no sense of abandonment on the HoN forums which you can sometimes find on ESR or the QL forums - the community is always informed on what S2 are working on, what bugs they are aware of, and any controversial thread will have a few in-depth replies by a S2 developer. I do not mean to deride Sync for the job that he has been doing, however S2 have been a lot more motivated and active than iD have been in this regard.
The last big difference is the frequency and size of updates on either side. S2 have been releasing massive patches just about every 2 weeks, almost always improving on the data structure, interface, gameplay/balance and content of the game. iD have been releasing sparse, scant updates, and although they have implemented 95% of the things the community has been bitching and moaning about, all updates seem to have been met by a sense of "is that all?". Here again, S2 have the advantage of not having to deal with pedantic web development, and I believe they have a larger, more motivated team due to having a set income from people pre-ordering the game, rather than a risky ad-based revenue stream.
In the end I have a lot of respect for what both projects have done. Co-ordinating a cloud based system with a master server as seamlessly as either have done it is not as easy as all the players seem to take for granted. Have id failed with quakelive? I don't think so. The product as it is today is very good. It is unfair to compare their progress with S2's larger team, who seem to have gone all-in on the HoN project. I do think there are a lot of lessons for id to learn from this project, and I just hope they can carry them forward to the next Quake, for the true revival of deathmatch.