4 ms·

The Google Wave grave reminds me of what a weird non-product that was. Has anyone done a good post-mortem/retrospective on that? Always seemed ahead of its time

The Google Wave grave reminds me of what a weird non-product that was. Has anyone done a good post-mortem/retrospective on that? Always seemed ahead of its time on a lot of levels, especially technically (one of the few times I've ever seen a single tab crash all of Chrome).

waterlesscloud 14y ago
Wave is one of those things that some future product will play off of and be massively successful, and we'll all point back at Wave as the forgotten root.
glabifrons 14y ago
I hate posting something like this without a link, but I read an article quite some time ago (supposedly) written by one of the core developers of Wave. To massively oversimplify the article, he blamed it on having way, way too many developers working on it at once... lots of inefficiencies crept in as portions were duplicated, it ended up much more bloated than it should have been, etc. IIRC, he also acknowledged many other contributing factors, like the slow release of the features presented in the original videos (some of which never were released), people simply not knowing what to do with it, etc. Something else I remember was that he suggested the code should/could have been a tiny fraction of the size and dramatically more efficient for what it did. I'm not saying I agree with the article at all, it's just what I read. For all I know it could have been a hoax. This appears to shed similar light on it: https://news.ycombinator.com/item?id=3101201 https://news.ycombinator.com/item?id=3101201 Personally, I loved Wave and was hoping to run my own server when they released what code they did release. Unfortunately, what has come of it is still lacking key features I'd want before doing so. (I'm not a Java developer, so would not be able to contribute)
- mark_l_watson 14y ago
  Give Apache Wave a chance. I like the slightly simpler and cleaner UI and it is easy to install and run.
sesqu 14y ago
As I recall, Wave suffered from a poor reference implementation and such complexity that amateurs were unwilling to create their own. It also received a lot of bad publicity from branding and people who were unsure about what to make of it, but what killed it was that it never left beta (much more concretely than other Google products).
takluyver 14y ago
I was a Wave enthusiast. I felt like not even Google really saw how the platform could have been useful. They presented it as a communications thing - e-mail and chat mashed together. But you can edit other people's messages, which makes communication really confusing. Wave had the potential to be like Google Docs, but using an open protocol for live collaboration, instead of locking you into Google's services. So you could host it yourself, and seamlessly share it with anyone using a compatible system. In my more cynical moments, I wonder if Google realised this and killed it before it could compete with Google Docs.
- mjn 14y ago
  I actually wrote one collaborative academic paper using Wave before it disappeared. Was interesting. Better than Google Docs in some ways, easier to break things into pieces that could be commented on and moved around.
spinchange 14y ago
I saw it as a hybrid between document collaboration and a message board, but they hyped it into the stratosphere as a communications medium & social thing too. Then everyone had to wait months (*edit actually a year or more, IIRC) for invites that came out painfully slow which had the effect of building anticipation and expectations even more. When it was finally widely available, most of the people I know were disappointed that it wasn't a social network, it was a weird document collaboration-message board hybrid that didn't use documents, posts, or any conventional metaphors and therefore didn't live up to its year of hype and anticipation building.
jayflux 14y ago
I really do think wave's only problem was it was just ahead of its time. The idea behind it was good and it worked well, yes it heavily used js resources but faster JS engines these days would cope with that. Its a shame really, but I supposed they morphed the 'liveness' of it into Google Docs I'm not sure Google gave it the chance it needed.

waterlesscloud 14y ago

Wave is one of those things that some future product will play off of and be massively successful, and we'll all point back at Wave as the forgotten root.

glabifrons 14y ago

I hate posting something like this without a link, but I read an article quite some time ago (supposedly) written by one of the core developers of Wave. To massively oversimplify the article, he blamed it on having way, way too many developers working on it at once... lots of inefficiencies crept in as portions were duplicated, it ended up much more bloated than it should have been, etc. IIRC, he also acknowledged many other contributing factors, like the slow release of the features presented in the original videos (some of which never were released), people simply not knowing what to do with it, etc. Something else I remember was that he suggested the code should/could have been a tiny fraction of the size and dramatically more efficient for what it did. I'm not saying I agree with the article at all, it's just what I read. For all I know it could have been a hoax. This appears to shed similar light on it: https://news.ycombinator.com/item?id=3101201 https://news.ycombinator.com/item?id=3101201 Personally, I loved Wave and was hoping to run my own server when they released what code they did release. Unfortunately, what has come of it is still lacking key features I'd want before doing so. (I'm not a Java developer, so would not be able to contribute)

mark_l_watson 14y ago
Give Apache Wave a chance. I like the slightly simpler and cleaner UI and it is easy to install and run.

sesqu 14y ago

As I recall, Wave suffered from a poor reference implementation and such complexity that amateurs were unwilling to create their own. It also received a lot of bad publicity from branding and people who were unsure about what to make of it, but what killed it was that it never left beta (much more concretely than other Google products).

takluyver 14y ago

I was a Wave enthusiast. I felt like not even Google really saw how the platform could have been useful. They presented it as a communications thing - e-mail and chat mashed together. But you can edit other people's messages, which makes communication really confusing. Wave had the potential to be like Google Docs, but using an open protocol for live collaboration, instead of locking you into Google's services. So you could host it yourself, and seamlessly share it with anyone using a compatible system. In my more cynical moments, I wonder if Google realised this and killed it before it could compete with Google Docs.

mjn 14y ago
I actually wrote one collaborative academic paper using Wave before it disappeared. Was interesting. Better than Google Docs in some ways, easier to break things into pieces that could be commented on and moved around.

spinchange 14y ago

I saw it as a hybrid between document collaboration and a message board, but they hyped it into the stratosphere as a communications medium & social thing too. Then everyone had to wait months (*edit actually a year or more, IIRC) for invites that came out painfully slow which had the effect of building anticipation and expectations even more. When it was finally widely available, most of the people I know were disappointed that it wasn't a social network, it was a weird document collaboration-message board hybrid that didn't use documents, posts, or any conventional metaphors and therefore didn't live up to its year of hype and anticipation building.

jayflux 14y ago

I really do think wave's only problem was it was just ahead of its time. The idea behind it was good and it worked well, yes it heavily used js resources but faster JS engines these days would cope with that. Its a shame really, but I supposed they morphed the 'liveness' of it into Google Docs I'm not sure Google gave it the chance it needed.