Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Review my startup, colaab - A Silverlight 2 RIA for rich, real-time collaboration
30 points by stormideas on Jan 12, 2009 | hide | past | favorite | 95 comments
Over the course of the last year or so I’ve been spending all of my spare time (and increasingly more and more of my working day!) on building a RIA in Silverlight for rich, real-time collaboration over a range of resources:

http://www.colaab.com

Allowing you to comment and annotate documents, images and videos.

Inspirations include David Heinemeier Hansson’s talk at Startup School 08 (http://bit.ly/490Tkt) and 37 Signals in general. My dream is to combine their approach to monetization (build a service people want, charge a price for it, make profit) with a dedication to a richer user experience that is enabled by UI technologies like Flash, Silverlight, JavaFX etc.

The company and product has been bootstrapped up to this point, with contract work paying the bills up until now (and for a while yet I would think!). In terms of revenue, we’re operating a subscription based / freemium model.

Everything is hosted on Amazon Web Services (EC2 / S3) so hopefully that will enable us to scale up as demand increases!

I’m based in Edinburgh, Scotland and believe the time has never been better for small teams of motivated individuals to build truly innovative and user friendly software, which with the distribution and viral nature of the web, have a chance of being adopted by a global audience.

I appreciate that a full-screen Silverlight 2 application isn’t everyone’s cup of tea, but if you’re interested I’ve setup an invite code:

HACKERNEWS

Which will give the first 100 people to use it a free "Starter" account (worth $24 a month) for a year.

Thanks,

Bob

-- Bob Thomson storm ideas http://blog.stormideas.com http://colaab.com twitter: movingforwards




Ok. I spent 10 minutes to download SilverLight and restarted my browser just to view the video. Why don't you just use the flash video player?

Then, when I proceeded to the sign up page, I were "amazed" that you created the price plan with SilverLight too. And, it took time to load again just to view the price and to login.

When I stumbled upon the sign up form, there were too much information that I have to give. This doesn't really favor any users who want to try the service (first name, last name, email, username, etc in SilveerLight). And, I have to wait for a confirmation email to get started?!? Luckily, the confirmation email reached faster than I thought and I gave my actual email. I wish that you won't spam my mail box.

And, finally, I cousd get started with the service. I stared at it for 2 minutes and had no clue how to use it. Then, I clicked the close button. I will try the service again if someone in HN give a good review about it.

I know that it is partially my fault because I didn't watch the 8mins long intro video and the popup tooltips in the working platform.

http://riastats.com/ < You can see that there is only ~15% of penetration of silverlight on internet. Hopefully microsoft will put in more marketing budget in Silverlight despite the "failure" of Vista.

Sorry if I am too harsh on my language and I am not trying to discourage you at all.


Thanks. The reason for using Silverlight for the video is to encourage people to get over the Silverlight installation hurdle as soon as possible.

I've been back and forward on whether that's a good idea, and whether we should replace the demo video with a flash version. Will definetely think it over again.

We're currently redeveloping the sign up process as a more standard html form, clearly this was me getting a bit too swept up in my Silverlight development efforts!

I'm also working on a "Try it now, in a demo workspace" feature to allow people to try it without submitting any information at all. Maybe I should also change the email confirmation step to be something that needs to be completed in time, rather than something that needs to be completed before you can log in.

Do you think 8 minutes is too long for the intro video? Should we make a shorter / snappier version?

Thanks for all the feedback, very useful.


I think you're more likely to lose people immediately with that hurdle - I certainly wouldn't install Silverlight just to view a video to see if I was interested in a service, but if the video made the service look interesting enough I might well download a plugin to try it.


Seems obvious to me now! Will sort it out.


I was more than happy to download the Silverlight plugin (I didn't have it installed). After an apparent successful install (latest OSX) and restarting Firefox, I'm told the plugin's not installed. I don't know how typical this is but perhaps Silverlight's not ready for the prime time yet?


> I've been back and forward on whether that's a good idea, and whether we should replace the demo video with a flash version. Will definetely think it over again.

Do an A/B test for a month using Google Analytics. That should settle it. Anything else is pure speculation.

This is simple but very, very important : any testable hypothesis should be tested. Long live popperianism.


A/B testing is definetely the way to go, my only excuse is that up until now I've had my head so deep in the development to even begin to think about much else!


I dont think 8 minutes is too long, but make sure you have a bullet list that says what your app is, what it does, who its for. Make it simple, make it exceptionally clear what you're doing and why your customer NEEDS your service. Re-iterate how bad collaboration is without your software.


Was the confirmation email in silverlight too? ;)


Nope.


extracting some constructive criticism from this (and my own thoughts:

* make the intro video in flash as well as silverlight! :)

* I'm personally really against confirmation email things.

* price plans/intro as HTML/PNG/Flash as well as silverlight

I think its important to really sell your site to users without silverlight - they will absolutley install silverlight if they WANT your service.


I think you're right - allow them to evaluate the service with as few hurdles as possible first and then deal with the Silverlight install issue once they have made a decision. Working on it already!


If it took 10 minutes for you to download the plugin you might want to try replacing your 14.4k modem.


Jeez, you're modding me down for questioning a 10-minute download for a 4.6M file?

What is this? 1995?


There's lots of reasons not to want to install a potentially bloated "platform" like Silverlight besides download speed.


"Was that really how long it took to download the file?" is questioning his claim. Telling him to replace his 14.4k modem is unnecessary snark with a hidden question.


He's being a wiseass. It's not even funny.


Well to start, rethink the EC2 web serving, you're spending more money than you need to. Consider linode or slicehost as bandwidth is way cheaper than amazon.

Second, why silverlight? The whole idea of online collaboration is allowing as many disparate people come together regardless of device. So you've thrown up roadblocks to two platforms that can no longer be considered marginal (OS X and Linux) for what is an arbitrary technology choice. At least with Flex you'll hit 98% of it and you can't sell me that silverlight has any advantage over Flex that is worth that.

I'd love to give you more feedback because I'm very interested in online collaboration, but I can't because I'm on Linux and won't install SilverLight on my mac.


This is a strong point for a bootstrapped company. We used EC2 for 2 months and then switched to Slicehost (and we'll try linode one day). That reduced our expenses a lot and we have the same quality of service. Think about it!


Thanks, Will check it out.


I think silverlight could be an issue, but plenty of companys get by on windows only tools.

Why can't you install Silverlight on OS X? I had to quit my browsers then run the installer. It failed if they were open.


I was able to install silverlight on OS X and FF. When I clicked to watch the demo it asked me to install and I clicked the link which downloaded 2.0. I didn't even have to restart my browser when it was done.


Why should I go through the hassle to evaluate something I won't potentially use? Even the intro video is in silverlight. Perhaps if it were in flash I could make some determination if it's worth the hassle.

And what's the point of the hassle in the first place? Someone has yet to explain why SilverLight is better for me as the consumer than Flex/Flash.


Since when has HN stuck with the defacto technology...

Ok, this may age me some. But people said exactly the same thing when Java and Flash first came out... if you go back far enough they were also intrusive annoying frustrating bug ridden technologies.

I'm not directly supporting SilverLight. I'm just supporting as many competitive web platforms as possible. They all add something, even the evil Microsoft ones. That being said Microsoft is losing the battle with that horrible SilverLight install. It needs to be as easy to install as Flash. It should be as close to transparent as possible.


>> (Java+Flash) "if you go back far enough they were also intrusive annoying frustrating bug ridden technologies."

To a fair number of people, they still are.


True enough!


I think having a flash tutorial (as well as a silverlight one) is a great idea.


or, just a flash one, since it's what is commonly used for videos.


That's sounding good to me. Going to add detect for Silverlight and switch in a flash version of the intro video if Silverlight isn't present.

Thanks!


I installed it so I could watch Netflix content instantly


Great job! I've only looked at the video though. Your user experience looks quite intuitive which is important. I would continue to work on that. For example, those that say to use Flash for video, well you can detect what the browsers capabilities are.

Don't worry too much about the Silverlight installation. If the product is good, that won't be a barrier.

I also believe RIA apps are much better than AJAX based apps.


Thanks, appreciate that. Very nerve wracking pushing something out there to the public that I've put so much time and effort into!

The user experience is meant to be one of the key differentiators and I'll continue to try and improve it as time goes on and feedback comes in from users.


I've had a quick play and set up an account.

There's a few usability quirks that immediately sprung to mind;

I didn't find it particularly intuitive to actually dive in use:

1. When you first set up account and start a workspace, the first thing it asks you to do is choose the people you want to share it with. This seems "wrong" to me. I would expect to be able to set up a workplace, add documents and wotnot and only then be bothered about who I wanted to invite in.

2. I then tried to work out how to add an item to a workspace... there were buttons on mouse over for Manage, Share and Delete - so I sat there trying to find 'Add document'. It wasn't until I eventually clicked on the workspace that I 'entered' it. I wasn't really expecting that 'enter to do something' metaphor. How about a mouse-over 'click to enter message'.

3. Add video resource - how about some clue as to the formats you accept. None that I tried. that's for sure. It's a bit distracting to see the Web page URL input go scrolling by on the way to the file chooser.

4. Thumbnails of the documents would be good.

5. The Black pallette really isn't to my taste.

6. The big one... without the ability to actually edit Word or Excel files (as opposed to annotate), it isn't really a collaborative tool in my opinion, I'll probably keep using Google Docs or Drop Box and I suspect others will too. Sorry, it's a lovely idea apart from that. I'm not sure I can even see myself using it with the free account.


Angostura, thanks for the feedback.

1. I'm aware that the initial create workspace wizard is too complex. You can do what you suggest by inviting no-one, adding your content and then editing the workspace to invite the users but clearly that's not intuitive enough at present. Will try to improve.

2. Hoist by own petard here! I surfaced some functions to the list view in order to try and reduce / simplify the interface once you were inside the workspace. Will add rollover as suggested.

3. I'd love to support more video formats in time - Silverlight natively supports .WMV files so this is what colaab supports presently. In the future I'd like to broaden this out and perform server side conversion for display inside the application. The point about the way the wizards move stages is noted, will see if I can improve on that.

4. Thumbnails for all resources is in the JIRA features list, will put a note on it saying someone else requested it. At some point I had to stop implementing features and get a beta version out there and unfortunately this was one of the features that didn't make it - it's in the works!

6. Editing is such a huge area, and while I'd love for colaab to provide some additional editing capabilities with the resources I have available it's just not possible. For the moment I am hopeful that the need to collaborate in this way (comments and annotations) over a wide range of content types will fit with enough of a need to get up and running with an initial user base.

Many thanks for your feedback, will see what I can do!


Glad they were at least marginally useful.

1. Perhaps the Wizard should initially ask 'what would you like to do first - add resources, or invite people to the space'?

2. I'm not sure that adding a roll-over is the correct idea here, so I don't want to send you down a blind alley. It may be the correct solution is to actually avoid surfacing the functions at all (they still have to be provided inside the workspace, I presume so you aren't reducing complexity).

Instead maybe it is worth looking at how you could amend the workspace design to make it look more 'clickable to enter'.

3. Understood. All I would suggest is you list the supported format(s) to avoid Mac users like myself attempting conversions blind.

6. Completely understood. The comment was a bit brutal but I thought it was worth making.

The very best of luck.


Very much so, I really believe in the idea of putting something out there, gathering as much user feedback as possible and then iterating quickly as a way of developing software that meets the needs of users.

Obviously so much easier to do when dealing with a SaaS style web application than a desktop one!


It looks very cool, I just created an account. Not sure if its usefull for me, but it is at least interesting to see a nice silverlight app. Did you build it with Blend or just Visual Studio? Personally I would pick Flex due the market share it got. But silverlight is much more enjoyable to develop with. Mainly because the adobe builder is not as good as Visual Studio and Blend.


While I love Visual Studio, I have to say that Flex Builder 3 is very nice, just as nice as VS in my opinion. What do you like more about VS that Flex Builder 3 lacking?


Thanks! It was built using a combination of Visual Studio and Blend. Tend to use blend to do a first pass at UI elements then hand crank things in VS from then on.


It's crashing both IE8 and FF3 when I attempt to type the first character in the email address field on the login page.

Silverlight version 2.0.31005.0.


I've got an open support call into Microsoft about this one, not to shift the blame but it seems to be a bug with the Silverlight plugin rather than something we are specifically doing.

Thanks for the feedback!


Looks great! I tried uploading a pptx and it worked flawlessly. This is probably the most impressive demo I've ever seen on HN.


Thanks! That's really encouraging. Glad you had such a positive experience.


This is a general point of view but I don't see the perks in using any RIA technologies (Flash or Silverlight) when javascript librairies are available everywhere (and I'm talking about jquery). What can you do with silverlight that can't be done with jquery?

I didn't try the app (because of silverlight) but I really like the design. Is black the new blue?


For me personally I was able to build colaab faster and more enjoyably using Silverlight rather than jQuery (which I love and use a lot in other projects).

Of course this is of no matter to the end user, but where it does have impact is that hopefully I was then able to spend more time working on features that add value to the application, thus benefitting the end user.

There's also some area of the application (like commenting on videos) which would have been really hard to develop using just javascript.

In terms of the design there's definetely been a trend towards darker colours. I love the style that Adobe have used for products like Photoshop Express and this was definetely an inspiration.

Thanks!


That's a great answer. I like the "I'm working with the best tools I know" instead of "I'm working with the latest tools to be part of a trend".

Working with tools with which you're the best makes great applications... as long as you keep updating your toolbox.


• Silverlight could be an issue, the adoption rate isn't that great so the majority of visitors would have to install a plugin to be able to do anything — people get turned away from sites because of a slow loading page, so being prompted to install a plugin may be enough for them to bounce back to where they came from. Plus of course a lot of corporate environments have a ban on installing new software, and a lot of mobile devices (iPhones, etc.) don't even have Flash, so Silverlight is a long way off...

• Black — arrggg! Personally I hate websites which have a colour palette ranging from #ccc to #000! They even disorientate me a bit…I think! It may just be me, but somehow I doubt the rest of the world loves such a dull colour.

However it's a very impressive app, you can tell you [guys (?)] have put a lot of work into it, and I wish you the best of luck — if I needed such a product I'd be MORE than willing to pay for Colaab! :)


I'm hopeful that the Silverlight plugin issue becomes less and less over the course of the next few months / years. Microsoft are pushing hard for adoption so (for me at least!) fingers crossed.

Already had some great suggestions from HN on ways to improve in initial experience (flash intro video, html sign-up form).

Shame you don't like the black but glad you found the application positive - I've put almost every spare waking hour into it over the last year, so hopefully there's enough people out there who share your opinion (about paying for it) for it to become a viable business.


wow, really impressive application (I made an account)

I couldn't get "website" resources to work? when I load them it just fails, odd!

the app wasn't full screen for me by the way, merely full window :)

How did you find Silverlight to develop in? are you on twitter? you should be :)


Thanks, really glad you liked it!

I'll take a look into your issue with the website screen grabs and get it sorted.

Yes, full window is indeed the correct term I should be using.

To be honest I found Silverlight a joy to develop in. In the past I've tried AJAX and Flash / Flex when trying to develop this kind of RIA application.

Personally I don't enjoy writing code in Javascript, and found it difficult to bring my ideas of how I wanted the user experience to come together into reality.

With Flash / Flex I could get the richer experience I was looking for but was then developing in a different application / language when working on the UI component as opposed to the back end and this was less productive.

Thanks,

Bob

p.s My twitter username is "movingforwards".


cool, I proded the european amazon web services dude to check out your site - nice publicity :) (he's @simon on twitter, I'm @plc)

Maybe, MAYBE, a little too many refelctions on the site. Just my 2 cents.

Can you turn this web app into a desktop app easily with silverlight?


At the moment if you want to turn it into a desktop application then you've got two choices:

1. WPF application - copy / share the code with a full blown WPF application and fix any incompatibility errors (Silverlight is a subset of the full .NET framework so there shouldn't be many).

2. Mesh Application - Live Mesh applications include the ability to run Silverlight outside of the browser, like a desktop application (short cuts, start menu items etc). I've done a bit of a proof of concept of this (http://www.vimeo.com/2542327 for an overview) but this introduces a hard dependancy on the user having Live Mesh installed so for the moment it's just proof of concept.

Thanks.


do you run OS X? I guess you could just make a browser app (eg, Fluid)

I was just thinking users like having a "desktop" application - might re-iterate the lovely Silverlight experience if the user could "download" a copy.


I would like to be able to select the text in the intro boxes, and pretty much anywhere else that isn't a button.

Also, I uploaded a pdf -- when looking at documents it's hard to read I'm running @ 1900X1200 and it's difficult to read; I'd work on the interface -- also, I assume you convert the pdf to an image? Why can't I select the pdf's text?

Other than that it looks like a good start, and silverlight seems like a good choice.. I recently chose flex for a project, but almost chose silverlight.


  * It's way too shiny. Like by a factor of a billion.
  * I can't find somewhere to just click "Free - go - try it"
  * And I have to install silverlight??? You've just reduced your potential market to a very small niche.


I think the intro video is too long. For the intro, skip the sign up part, just get right into it. Try to get it under 5 minutes. And invest in a good microphone. It sounds like you're talking into a tin can.


My thinking was to show people how quick and easy it was to get going from sign up through to actually using the product but perhaps this might be better suited to an instructional video rather than the main promotional one on the homepage.

Will order a microphone and see if I can get better quality sound on it, thanks for the shout. I'm normally too busy squirming at the sound of my own voice to notice the quality of the recording!


I have to second the comment on the length. In particular, I think the first minute where you show yourself typing in your name to create a new account and then checking email to verify the account, etc is a little unnecessary. At this point in the video, I haven't even seen the product so I'm definitely not sold on creating an account on your site or watching a movie about it. Maybe you can move that part to another video for people that might need help creating an account.

It would also chop off one minute of an 8 minute long demo.

It's cool that you decided to go with Silverlight despite the fact that it's less tried than Flash. Good luck!


Very impressive. I tried a word doc and a web page...nice UI.

I'm curious about the architecture. Did you use Model View ViewModel?

How are you converting the word docs? Do you convert every type of resource to an image?


Thanks, glad you liked it.

I started moving to MVVM as the project progressed and I began to see the limitations / issues with using UserControls and code-behind.

To be honest the application grew from just hacking things together initially so there are some older parts that were just me picking up the technology.

As time progresses I'm trying to move more and more of it to MVVM and get everything covered by unit tests - at the moment the majority of unit test coverage is on the server side - the client is a little light in that department atm.

Word document conversion is a nightmare, and is some of the more obtuse code I've ever written. At the moment we convert to XPS and work from there but I'm sorely tempted to go to images and not loose too much sleep over the fact that you then get pixellation as you zoom in a lot.

Thanks for the feedback, much appreciated.


No one wants to install yet another unused technology (Silverlight) for an unknown new product.


I took the decision to develop using Silverlight becuase I've got a strong background in using the Microsoft stack for development, wanted to get in on the ground floor with a new technology and follow it as it matures and becuase I wanted to (try!) to produce a richer user experience within the browser.

As I said, I understand that this is going to be a limiting factor in some regards, but I couldn't have built what I did in any other technology (that's not to say that someone else couldn't have, just for my particular circumstances).

I'm hopeful that as Silverlight adoption increases this will become less of a problem, I'm also hopeful that colaab may act as a showcase for my Silverlight abilities and lead to further contract work, which in turn can help to fund product development.

Thanks for the feedback!


Silverlight will be bundled into all versions of Windows from now on so you won't have to worry about adoption.


Will that work in firefox/safari/opera on windows?

Also, the web isn't composed of windows machines. It's a mishmash of a billion different devices. The vast majority of which, know html, javascript, and not much else. Seems a bad idea to only write for a small subset of them.


Well, I'm a Mac user running Safari. Do I mind installing a free plug-in to get access to an interesting app? No.

I don't see where you (necessarily) get "small subset" from.


You don't mind potentially making safari crash more? You don't mind adding to complexity, disk space used, memory usage, potential vulnerabilities?

I do. The less "installed" software I have to go wrong, the better.


I don't mind making Safari 'potentially crash more' - I only mind if it actually does crash more. Since Safari crashes are very few and far between, and haven't noticeably increased since I installed Silverlight some time ago, this potential problem isn't high on my list of priorities.

I suppose I can always go back to Lynx if these things really bother me too much.


I've used Silverlight on Firefox (Vista, Windows 7) with no problems, not even having to restart the browser.

Silverlight is actually pretty good; it's a much easier experience for developers than Flash/Flex. You might hate Microsoft, but you shouldn't hate the tech.


it's hard not to hate the tech when historically MS makes their stuff ignore standards and evolve into something for Windows only.

Adobe, while not perfect itself, has been committed to keeping Flash cross platform. This makes Flash harder to hate than Silverlight despite the high cost of developer tools


IE is still 70% of traffic. So, if you write something in Silverlight, and IE maintains that, you're going to work in 70% + however many non-IE clients have Silverlight or are willing to get it.

That will be most everyone. People will do just about anything you ask them to. The prevalence of phishing has proven that. They'll put their online banking creds into any site that has bankofamerica somewhere in the domain name. They'll buy Viagra from an anonymous email through a Costa Rican pharmacy. They'll most certainly click the button that says "Install Silverlight from Microsoft to watch the video".


If using Silverlight helps you do things that would have been harder otherwise, great.

But the other reasons are silly unless someone is paying you to increase Silverlight usage.

The difference between a biz and a hobby is that biz try hard to avoid doing things that don't help the bottom line.


Where do you live? I'll hire you ;)


It takes a few startups to jump in and push Silverlight into popularity. IMHO, it is a better and more open RIA technology than Flash, so I'm happy to see it used! (Although, you don't have to use it everywhere. The signup form didn't need it, for example).


It only takes MS bundling it to make it popular on the client end.


Aha, yes. I got a bit carried away there. Currently redeveloping in ASP.NET MVC as a more standard web form!


>IMHO, it is a better and more open RIA technology than Flash

A lesser evil is still an evil.


What does that even mean?


Silverlight being better than Flash doesn't logically conclude to making Silverlight a good idea and deserving of support. A lesser evil is still an evil.

I consider the fact that Flash is pretty terrible a feature, as at least is helps discourage people from only using it in places where it's absolutely essential. I firmly believe RIAs are a fundamentally bad idea and the quicker they are replaced with open standards the better.


Interestingly, I have silverlight installed on my macbook (Until later today probably when I uninstall it).

The reason being, earlier this month I bought an xbox360 (There is no choice, be stupid and spend a ton on a ps3 with its ridiculous bluray, or be stupid and spend less on an xbo360).... I bought the lips game, and went to lips.com to see what songs you can buy. It requires silverlight... for a list of songs. I installed it reluctantly, and in the most hideously badly rendered font you can imagine, it said "Song list coming soon".


well I just did, so erm, yeah.


I believe in this case, "no one" = most people. You apparently aren't most people.


so if you found a great application that perfectly matched your needs, you wouldn't use it purely because it required installing software on your computer?


The real scenario is: Two companies make similar software that fulfills your need. One requires a download and the other doesn't. Who wins?


the one thats cheaper


Take "similar software" to mean all things being roughly equivalent, including price. Jeez.


That's also untrue. If one was cheaper, but required me to install some desktop software, I'd happily pay the extra to have the better solution.


I'm pretty sure I had to download silverlight to watch netflix movies on my Mac.


Netflix are using Silverlight for their streaming service, see http://netflix.mediaroom.com/index.php?s=43&item=288 for details.


Not a good example. Netflix would have chosen Flash probably if it wasn't for a special deal with Microsoft. They traded something in exchange for choosing an inferior solution (inferior in that it requires a download).


Yes. I'm betting against installable software.

The whole point of a collaboration type app, is that you can use it from anywhere - your laptop, desktop, iphone, nokia, netbook, internet cafe, library, wii, etc etc

Requiring a user install some software negates the whole point for me.


I'm also betting against installable software (Silverlight plugins excepted!). You just need to look at Joost's traffic since they went web-based rather than desktop client:

http://www.techcrunch.com/2009/01/08/joost-has-a-heartbeat/

I'm hopefull that over the coming months / years Microsoft manage to get Silverlight out there and available on as many platforms as possible, allowing me to write applications that can take advantage of that reach.

In the meantime the aim is to start building up enough user numbers to create a viable business, I've got very low overheads and no debt so hopefully that can come from the existing Silverlight install base + people who are willing to install it if they think the application will do a job for them.


Sure, what I could see of the app, and design looked very good. So congrats on that.

Good luck with it :)


I believe in this case "most people" = some people.


I tried signing up and have Silverlight installed. But it just hung at the sign up stage wanting me to install siverlight even though I already had it installed. I even reinstalled it silverlight to no avail.


Yeah, Silverlight. No thanks.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: