To me, Wave failed because it wasn't a product. It was an engineering exercise. There was a lot of interesting technology in Wave (eg the Wave protocol itself, federated servers).
But what problem did it solve?
No one really seemed able to answer that.
When I looked at Wave I saw a set of primitives from which you could build email, IM, Twitter or any other number of communication mediums. The problem is that people could already do all of those things elsewhere. Did Wave make it easier? If it did, it might've stood a fighting chance. But it didn't. The UI was complicated. On most browsers it was slow.
Part of what made Google the search behemoth that it is (apart from the obvious algorithmic improvements over what came before it) was the simplicity of it. It was the simplest of pages (compared the shotgun portals of Yahoo and the like) with a giant text box. Anyone could look at it and know what to do. What's more it was fast.
So I would take Wave as an object lesson in what happens when you seek to solve an interesting engineering problem rather than build a product.
For me, Wave had one very strong use case: collaboration and discussion intended to result in a persistent document. Forums, mailing lists, e-mail, etc. are great for discussion, but they are generally ephemeral. Documents are persistent, and with Google Docs can be created collaboratively, but the communication channels around them are weak. You're limited to distracting in-document text, limited comments, or entirely out-of-band communication e.g. e-mail. With Wave you could have the document, editable by anyone, and embed discussion in it and around it. You can discuss a point, updating the base document as discussion converges. It's a brilliant medium for group brainstorming and drafting.
While I definitely agree this would have worked as a great collaboration tool, anyone I tried it with just did not get it, and we ended up using it as a I'm-seeing-what-you're-typing chat box.
I upvoted you, but I disagree that Wave didn't solve a useful problem. In my opinion, it solved some key collaboration problems. The problem was that it was marketed as an engineering project, instead of a consumer product.
Google positioned Wave as a new communication platform that would replace email, and that was just too much for users to wrap their heads around. Nobody could find a good reason to stop using email and switch to Wave. Google should have targeted the much smaller use case of document collaboration, even if the technology was capable of handling more, just to show people how they can start using the technology right away without having to change too much of their existing workflow.
It's a classic example of the engineers letting their dreams get too big and forgetting to focus on the immediate problem they started with. Wave could have easily been saved by a better marketing team.
Ironically, it seems Google has too many smart engineers, and has lost touch with what the average user needs or wants.
We tried using Wave for our startup. Really, we did. We tried using it as a collaborative tool, as an communications tool, and as a toy.
The bottom line was that we found we'd rather share a Word doc via DropBox, or send an email via gmail, than try to use the clunky, slow, confusing interface.
In my experience it was the product that failed. YMMV.
It was slow and clunky, but that's something I forgave since it was still a preview and I assumed it would be fixed given more time. I'm talking about the core concept of Wave when I say the main problem was marketing. Obviously the product still needed development work, but it was on the right track IMO. It fit into my workflow very nicely, and I definitely miss it. I would love to see somebody take the open sourced code and release a more simplified, targeted version for group collaboration.
Cletus since you work at Google, can you share what the feeling was internally about Wave? Were people excited and sure of its success or was there some doubt?
I'm sure you can appreciate that I have to respect the confidentiality with which I've been entrusted as a Google employee. This includes not only commenting on internal matters I do know about but also matters which it may appear I have knowledge of. It's not in my interest or Google's for me to unwittingly start a rumour because of something I carelessly say.
Wave predates me joining Google. My thoughts on it predate Google (I think I wrote a blog post about it some time last year).
So, just to be clear, any views expressed here are strictly my own. I'm sure you understand.
Despite the hype... I don't think Wave failed. I blame all of 'us' (where 'us' is defined as alpha geeks, silicon valley startup folks, techies, etc) for this misrepresentation. We all this products must be wildly successful with millions of users or it has failed... we tend to see failure and success as this very black and white / binary thing... but it's not.
Google is probably discontinuing use of Wave because it didn't live up to their expectations and didn't have as many users as they need it to in order for it to be viable. If this were created by a smaller company, however, it could still be considered successful... I haven't been able to find any numbers on how many active users wave has but I'd guess somewhere around 50k - 100k... For any smaller company that'd be a huge success...
I'm really tired of reading all of the rhetoric that acts as if it's just oh so obvious that Wave isn't a good product. For 50k users it's a great product that they love... It doesn't need to have 2 million users to be a good product... Personally, I think Wave is a BRILLIANT piece of software and I use it exclusively for communication with my 6-10 close personal friends.
I have one gripe - the mobile interface is terrible... Maybe if this could've been fixed more people would've used it.
When you market to 100M+ people, you can find 50K-100K people to love almost any product. That does not make the product good nor a success for the company that created it.
If Wave were created by a smaller company, it would not have been able to garner that much hype and tech blogs ink and therefore wouldn't have attracted that many adopters. In fact, without Google's reputation, a small company creating such a product and protocol would be dismissed as too obscure and complicated.
I thought Wave was awesome when I tried it out with all my engineers friends. But then we somehow stopped, I didn't even remember how or when we stopped. Wave just seems like a really cool software that solves problem nobody has.
I personally opted to join the Wave team because it was the first time that I'd seen Google try something quite different. Google has many other cool products, but many of them are online versions of existing desktop tools, or Google versions of existing online tools. Wave was a unique blend of a variety of communication tools, and it intrigued me.
In addition, I was in a role where I spent half of my time communicating and collaborating with different groups, and after failing to be impressed with existing tools, I was particularly interested in seeing how Wave might make my role easier. And despite its various flaws, it did indeed make it better.
In the end, I'm happy that I joined the Wave team. It was a really interesting project to work on and a great learning experience.
The real problem for Wave was that it had a 20-60 minute learning curve, almost no one I knew was using it, and it was hard to convince people to use it because of the learning curve and no one else was using it.
Wave is going to be one of those things people talk about for years afterwards wondering why it failed.
In the end, it probably was two things: 1) Wave didn't fit into some neat category that people already had about stuff, and 2) Google wanted to own all the data
You are correct about #1 but #2 is way off. Where did you get that assumption? From everything that was said or done about Wave, it seemed like they have no problem with having the data be outside of Google's silos. If they wanted to control it all, why did they want to have it be federated and distributed using multiple variations of Wave or custom apps that speak the same protocol?
I contacted the development group at Google and was told the Wave server code was proprietary and they weren't giving it out. Yes, the protocol was open, and the idea was that there would be a lot of clients and such that would use it, but if a team wanted to take the server code as-is and start plugging in sensitive information? The data couldn't live in-house.
That's why several projects I had decided not to use Wave. Shame, really. I liked it a lot. I hear that the server will be made available under the Apache license, so I'm looking forward to that happening.
Google's own Wave servers probably plugged in to BigTable and who knows what other proprietary stuff. So the source code probably wouldn't have been so useful anyway.
More specifically, individual invites. They were a smart what to roll out Gmail, which could be used on an individual basis (i.e. I don't need all my friends to sign up before I can start using it).
But isolated invites to a collaboration platform? That is jaw-droppingly thoughtless. Had Google adapted the strategy to the product by ensuring that invites went to people who already worked together as a group (i.e. established partners on a research project, members of the same design agency etc.), then the fate of Wave may have been very different.
But by making such a galacticly dumb opening move, they left smart people wondering "what other major problems have gone totally unaddressed? And do I want to invest an hour? a day? an entire weekend? to find out? By myself?"
And with that, they lost all support from the early adopters willing and able to validate the concept, replacing it with a bunch of bad buzz (no pun intended) that poisoned the whole thing.
Definitely. I got a Wave invite and had no one to use it with. Each new Wave user should've received a few invites to send immediately to the folks they'd want to collaborate with.
It was a great product, but it became 'another inbox' to me. It was designed to make communication frictionless, but it felt like a drag to check.
New communication services need to provide big, obvious benefits over existing services, then communicate those benefits with early adopters in a way that trickles downstream.
If we can learn one thing from Wave, it's how deeply embedded existing communication methods are in our way of living and working; it's hard to pry business users away from email and social users away from Facebook even if you're Google.
In this sense, Wave wasn't a bad idea; it was a great idea ahead of its time. I suspect that communication methods that are natural evolutions of existing popular systems (e.g. fax machines, Skype) are likely to have higher adoption rates than methods that take brilliant leaps too far too soon.
I'm not sure that iterating more would have helped Wave. The basic concept of a live chat/document editing platform was too far removed from what people are familiar with, so finding a clear use for it was hard. I don't think Google could have changed anything in the product to fix that, although they could have iterated the marketing process to pitch it to whatever audience were using it most and spur uptake in that community, perhaps.
It would be interesting to see if a Wave-like collaboration platform would be widely adopted if it grew organically over time out of an existing Google product -- like Gmail.
Wave from what I read was developed for use internally at Google when Gmail was insufficient. I'd be curious to see what's being used within Google right now in the collaboration space. People who were once using wave, what are they now using?
Up until the day I left Google (last Friday), my team was still using Wave. We were a bunch of ex-Wavers though, so we were more likely than others to still be holding on.
I should point out that our internal version of Wave was different from the external version in at least one very important way: when a wave was shared with a group that you were a member of, the wave was sent to your inbox. Externally, you would have to set up a search to find that wave, so basically nobody ever read group waves. The idea behind the decision was to reduce inbox noise (particularly important in beginning few months), but that also meant reducing utility for group collaboration.
We were close to releasing proper groups support in the external version, so you could easily choose which groups to follow/unfollow inside your Wave inbox, but were discontinued before that made it out.
So, Wave was awesome for group collaboration, but unfortunately, we never gave the external world much of a real opportunity to witness that.
(And yes, there were other issues with Wave. Just thought I'd point out that one.)
That makes a lot of sense. My campus ACM group used Wave an enormous amount when the beta invites started going out, and we kept two organizational waves which mimicked the system you describe. One was a list of the screen names of every member, and the other was a list of waves.
When Wave came out I thought it was really cool and used it for several weeks. Then it wore off, we began using email for stuff we did on Wave, nobody was on Wave anymore, so it slowly died.
One of the major factors for me was the crappy, slow UI.
Also, I think Wave lacked CONTEXT. It was great for collaborative editing, but which kind? With friends about parties? With classmates about projects? With colleagues about code? With customers about products? It imported a few Gmail contacts from each group, not enough for either one to make sense in the long run. It had some great features, but not enough for either one [compared to email/Facebook/specialized webapp].
> One of the major factors for me was the crappy, slow UI.
For me, that wasn't a problem in Chrome, interestingly enough. It was indeed extremely slow in Firefox though. Maybe Google didn't have enough time to optimize it for other browsers before the novelty started to wear off.
Wave was an amazing product and set of technologies.
It was a release and marketing failure. Should have been in the Google incubator at least another 6 months before Alpha/Beta, and shouldn't have made the huge mistake of releasing a product to users with limited invites when the entire purpose is collaboration. And that's what the focus of its marketing should have been: collaboration. Group chat, forums, and wikis, here's your replacement. Planning a trip or party? Use Wave instead of Gmail. That's all it needed. Geeks want to hear about the cool technology in it, everyone else just wants to know what the hell am I going to use this for.
I'm obviously not privy to Google's internal deliberations, but I've always thought that Wave was a great product - for people working in large institutions like Google. If I had to regularly collaborate with people in different offices, and with customers and suppliers, to keep track of a variety of project state and planning documents, Wave would be an excellent tool.
Unfortunately, my colleagues are in the same room, and I already have other ways of communicating with my friends, so Wave isn't much good to me, and would probably never achieve mainstream success. It might have had a chance with corporations fed up with Outlook, though.
Personally, I've never been surprised at Wave's apparent failure. To me, it was no different than any other attempt at a "revolutionary idea" that didn't work later in another revision or iteration (I think Hacker News readers understand this very well;a poor analogy would be a Microsoft Tablet and an iPad). A lot of discussion about it's failure is a direct result of Google's status and the amount of resources required to come up with a product like wave.
When I consider Wave in retrospect, I really think they limped in on this . It was a grand experiment. If they really wanted this to succeed, I don't think they would have proceeded with the launch the way they did. Reading the Quora answer sort of jives with this.
I'm surprised no discussion touched upon the Anonymous response in the Quora thread. It's more interesting than "why Wave failed" and is indicative of the state of Google. It brings an additional perspective to the discussion around Google's (apparent) talent retention problems. Combined with other anecdotal evidence, it really look like Google is at a loss for how to continually motivate employees with their current organizational structure
Agreed. The most interesting bit of information from the thread is the structure of the Wave team and how the incentives may have contributed to the quality (or lack thereof) of the shipped product. Kind of like incentives on Wall Street rewarding risky behaviour.
Briefly (I need to run), when it was announced/described, I saw it solving a lot of problems I've encountered:
Submarining discussions and decisions (e.g. email's that don't include the whole group or that don't get shared with new members joining the group).
Excessive work to reconstruct discussions, events, decisions. (Again, "big wads of emails" being a typical example.)
Centralized, rigid access controls. (You have to go through systems administration to add somebody to a project. Sometimes, that in turn means management sign off. Quickly, you're back to copy/pasting crap into, oh, for example, email.)
Wasted time. For example, the last three points.
Lack of dynamic, group discussion separate from physical and chronological synchronization and separate from the need for productivity killing meetings. Some companies have (finally) caught onto internal bulletin boards and the like, for this. But then somebody has to set them up, add users and credentials, and centralize a whole lot of bureaucratic decisions that might better be pushed out to the teams involved.
Lack of a centralized, uniform project communication history.
Etc.
Anyway, Wave seemed promising. But the client Google offered up was unusable for most people. It was literally too slow, at first, on the client side. It reflected the worst of Google's trend to abandon discoverability in its UI. (I'm reminded of this in a smaller fashion every time I help a client's employee take the dive into their calendar product. I'd argue that Android has a fair amount of it, too.)
Also, early adopters would join and have no one to work with. Even when you could get someone else invited, most didn't want to tackle it. Growth stagnated.
I still think Wave' paradigm solves a lot of problems. Maybe Apache and third parties will have better luck producing a usable client.
P.S. There were also many niggles and problems with the UI, such as default decisions about account creation and account information sharing. The UI was, basically, quite underdone. Finally, Google didn't continue to pour effort into it nor give it enough time to take off.
GMail's invite-based beta worked because there were no network effects in the product's value proposition. Even if those with whom I frequently corresponded did not use GMail, I benefited from the product. Wave was totally different -- everyone with whom I wanted to collaborate (at least in a particular context) had to have accounts, and I didn't have that many to give out. The product should have been implemented so that email/IM/whatever were used as fall-back "protocols" for those who had not yet adopted Wave. Also, the invites should have granted one the right to create N waves involving, say, 50 people max.
It can also be viewed in the context of recently posted (don't remember who's) article which analyzed RIM's "interactive pager" vs. TIVO's PVR as example of marketing positioning of "extension of existing product" vs. "a product creating new product category".
In this case it may have been more successful if they developed and marketed it as an "live collaboration" extension to GMail or as interactive collaboration layer on top of Google Docs.
So true... Wave was actually pretty damn cool in a lot of ways, but they seemed to do little or nothing to explicitly position / market it. Coupled with the fact that the original UI was a bit, erm, awkward, and we get what we have here today.
Thankfully Wave lives on at the ASF and hopefully people are going to be able to use the technology to do some mondo radical stuff.
Google wave was classic groupware ( see http://www.jwz.org/doc/groupware.html ). And I think the people at Google got confused in the sense they used it at a place they love (Google) so they thought it was great (Wave).
> Because it solved problems (granted, all of the easiest and least important problems).
Unifying the benefits of email and IM in a single conversation isn't, for me, easy or unimportant.
I thought Wave was engineering led rather than user led, but with some polish and integration into existing apps (like GMail and Outlook) could have been a killer.
If the flop, they question why you even started in the first place.
Wave was spin to try and reinvent internet communication. It wanted to merge email and IM and all these other communication forms together. Perhaps it was way too ahead of its time. Perhaps the execution was not as well thought out as it should have been. But the overarching concept is not ridiculous.
All I want Wave to be is Google Docs + Gchat on the same page. Possibly with the ability to link parts of the doc with the Gchat conversation going on when they were written.
Wave actually shows the other person typing and editing. Really cool. Skype doesn't do that, does/will Convore?
I use Wave and the main problem is the load time, and basically speed. They need to get some HTML5 and local storage into it, maybe some Node.js.
Also, Waves can seem to go on and on, and one forgets why a wave was created in the first place, and when one should be ended/closed.
So there should be a splitter button to start a new wave within an existing wave.
It's rather like a text chat with someone. You never know what's going to come up, so how can you categorize it before it's even begun? Or if you do, and it goes off-topic, then what? Do you head-butt a wall or do you go to the beach?
IMHO .....because it was,
the real problem was that people didnt see its value, scalable online collaborative editing where some of the editors are programmable bots, i dont think people really realised its potential sadly.....
I'm not sure why people are anti wave. I still use it for the purpose of multiple collaboration and it serves that purpose well. I honestly do not think it is "ultimately useless".
Just because you aren't a fan of a particular technology doesn't mean that you are actively "anti". I actually went back and watched the original Wave demo on Youtube just so I could reconstruct my original reactions, where were basically:
- That looks really cool
- I wonder how they do that (the synchronized live edits)
- I can't really think of any scenario where I would want to interact like that
Now I can see how some people might like it, but I really didn't - and I still find it a struggle to see it being used.
Perhaps the gulf between email and instant messaging isn't just a technological one, but actually corresponds the way most people are happy interacting?
didnt know that people can collaborate in IRC on one document and see real time changes going into a shared piece. The wave bots had also had the ability to collaborate, See http://www.open.ac.uk/blogs/telstar/2010/01/14/irm10-help-me... of a wave bot helping in a realtime environment to facilitate refrerences for research projects...... I think people that think wave was just a replacement for email, irc twitter are kinda missing the point, and google are largely to blame because that is the way wave was "marketed"[sic]
One can think of the entire IRC history as a kind of collaborative information resource. (A limited append-only doc.) And yes, there are programming groups who use the history in exactly this way. Some of them post to HN.
But what problem did it solve?
No one really seemed able to answer that.
When I looked at Wave I saw a set of primitives from which you could build email, IM, Twitter or any other number of communication mediums. The problem is that people could already do all of those things elsewhere. Did Wave make it easier? If it did, it might've stood a fighting chance. But it didn't. The UI was complicated. On most browsers it was slow.
Part of what made Google the search behemoth that it is (apart from the obvious algorithmic improvements over what came before it) was the simplicity of it. It was the simplest of pages (compared the shotgun portals of Yahoo and the like) with a giant text box. Anyone could look at it and know what to do. What's more it was fast.
So I would take Wave as an object lesson in what happens when you seek to solve an interesting engineering problem rather than build a product.