Introducing Bespin

As we strive to evolve the Open Web as a robust platform for application development, we believe in the potential for web-based code editors to increase developer productivity, enable compelling user experiences, and promote the use of open standards.

Today we’re launching Bespin as a project within our Developer Tools Lab to focus on this exploration.

Just as Mozilla enables massive innovation by making Firefox open on many levels, we hope to do the same with Bespin by developing an extensible framework for Open Web development. We’re particularly excited by the prospect of empowering Web developers to hack on the editor itself and make it their own.


Bespin proposes an open extensible web-based framework for code editing that aims to increase developer productivity, enable compelling user experiences, and promote the use of open standards.

Based upon discussions with hundreds of developers, and our own experience developing for the Open Web, we’ve come up with a proposed set of features along with some high-level goals:

  • Ease of Use — the editor experience should not be intimidating and should facilitate quickly getting straight into the code
  • Real-time Collaboration — sharing live coding sessions with colleagues should be easy and collaboratively coding with one or more partners should Just Work
  • Integrated Command-Line — tools like vi and Emacs have demonstrated the power of integrating command-lines into editors; Bespin needs one, too
  • Extensible and Self-Hosted — the interface and capabilities of Bespin should be highly extensible and easily accessible to users through Ubiquity-like commands or via the plug-in API
  • Wicked Fast — the editor is just a toy unless it stays smooth and responsive editing files of very large sizes
  • Accessible from Anywhere — the code editor should work from anywhere, and from any device, using any modern standards-compliant browser

View Introduction to Bespin

The Initial Prototype

As part of this announcement, we’re also releasing an early experimental prototype to demonstrate some of the concepts of Bespin and the possibilities that it opens up.

Bespin 0.1

  • Initial prototype framework that includes support for basic editing features, such as syntax highlighting, large file sizes, undo/redo, previewing files in the browser, importing/exporting projects, etc.


Bespin 0.1 Running in Firefox 3.0

Screenshots of Bespin 0.1 running in modern, standards-compliant browsers

All of the source code underlying the Bespin experiment is being released as open source software under the the MPL.

Get Involved

Mozilla Labs is a virtual lab where people come together online to create, experiment and play with Web innovations for the public benefit. The Bespin experiment is still in its infancy and just getting started. There are many ways to join the team and get involved:

Ben Galbraith and Dion Almaer, on behalf of the Bespin development team

91 responses

  1. hhxue wrote on :

    Good luck with the project.

  2. xlnv wrote on :

    I hope you can translate the chinese version in this tool.

  3. terkoz wrote on :

    could you make it downloadable some day so that I could test and code javascript php and HTML inside firefox?

    the problem with this is that you can’t test it and you can’t download what you’ve made.

  4. chalkos wrote on :

    This is just awesome, many features, great performance, and open source.

    Its visuals are very good.

    Thank you guys, keep up the great job. This is a rly good project.

  5. Eric wrote on :

    awesome love it!

    could you make it downloadable some day so that I could test and code javascript php and HTML inside firefox?

    the problem with this is that you can’t test it and you can’t download what you’ve made.

  6. Dan wrote on :

    @Dao – You know many blind programmers?

  7. Kalmi wrote on :

    I had (very) bad performance under ubuntu interpid using fx3 with an nvidia card.
    Installing driver version 180 solved everything.

  8. J Kenneth King wrote on :

    Not every problem that has a solution need be solved.

    I can’t imagine the amount of technical complexity required to build a web application such as this, but neither can I imagine a reason why it is better in a pragmatic sense than my current editor (emacs).

    Maybe I cannot see the same picture as everyone else, but I’ve summed up my initial thoughts at:

    Good luck with the project.

  9. mike wrote on :

    I would like blocks of code to be able to be tabbed in. This does not seems to work. Also untab with Shift tab doesn’t work. I would like text that is selected to be a different color. it is hard to read the text that is selected. Also control k cuts a line but control u should paste the line according to most text editors in linux.
    Great job on the editor. I would like ti to work on google chrome and opera. I have seen opera working fine with canvas on other sites before.

  10. Jorge Kobeh wrote on :

    I can only say: Wow!

    Keep up the good job. I’ll be reading your source code.

  11. Dmitrey wrote on :

    can I install it as Mozilla add-on via Tools->add-ons->get add-ons?
    Search for “bespin” doesn’t yield results.

  12. Andrew wrote on :

    a “publish to server” feature would be something I’d want up front, but it’s already incredible. way to go!!

  13. michael wrote on :

    Although I am impressed with this demo, I don’t like this idea of a graphic you can render text to on the browser. We’d all better say hello ‘drm protected’ textual content that can no longer be read by screen-readers, and content which doesn’t honour your font choices, or re-size properly to different displays.

    Already some news sites use some awful hacky javascript and layers to present uncopyable text, and it wont be long before they have embedded pdf-like viewers with completely custom renderers and unaccessible content – like flash but worse.

    Back on the demo, quite impressed – particularly that implementing the whole toolkit in javascript works at all, let alone at interactive speed. I was thinking of doing something like this in java (at least, the remote editor part), because of the access to a proper widget set and performance, but this shows that perhaps it could be done in javascript instead.

    Although a limited demo that works in english for able bodied people is one thing, a full-blown i18n capable toolkit, non-english or custom keyboard layouts, and full accessibility and the rest is another thing entirely (if it is even possible).

  14. Matthew Wilcoxson wrote on :

    Very nice. A fine example of what you can achieve with just JavaScript and HTML.

    And it looks to be a very useful tool too. I can imagine just going to a URL to update my website files. No need for file transfer, just quick and simple editing.

    The command line tool is also really nice. I can’t wait for the “set callaborate yes” command to work! (Couldn’t figure out the command to move between command line and editor though)

    I’ll keep my eye on this.

    And very nice scrollbars too!

  15. jdavid wrote on :

    I like the idea of a cloud viable editor. The power of this will come in when you can tie it to a source control tool like a wiki engine, that can cross link documentation right with code via syntax highlighting.

    so a function in example.js

    function somefunciton(var1,var2){

    //some set of operations

    auto generates a link around somefunction to a wiki page that describes some functions operation and lifecycle (changes, bugs, TODOs, and other meta data)

    As far as editors go, i think JEdit is the simple text editor to compete with, its on all platforms, and it has great plugin, macro and shortcut key support. Other text editors are not as cross platform, and since this is in the cloud, this needs to beat the cross platform options out there.

    Maybe this can compete in places of code cooperation where JEdit can’t, like a wiki and MS word compete with composing an online encyclopedia.

  16. bobinson wrote on :

    Sleek interface and I think now I can edit my companies html pages from anywhere in the internet. But, it looks like it doesn’t support PHP and as of now I couldn’t figure out a way to send the files / upload the files to my real server.

    another quick request will to add a command to open the files ie, open file

    also the command line should be context sensitive like Cisco router CLI


    project project_name should change the context
    and then
    open file_name should open a file in the project_name context


  17. bobinson wrote on :

    Sleek interface and I think now I can edit my companies html pages from anywhere in the internet. But, it looks like it doesn’t support PHP and as of now I couldn’t figure out a way to send the files / upload the files to my real server.

    another quick request will to add a command to open the files ie, open file

    also the command line should be context sensitive like Cisco router CLI


    project project_name should change the context
    and then
    open file_name should open a file in the project_name context

  18. Balach wrote on :

    freakingA… das ist zuper!

  19. Nikola wrote on :

    This is great! The only thing I’m wondering is whether or not Lando is going to help debug…

  20. Dmitriy Zasyatkin wrote on :

    I am all for a cross-platform IDE, but why not extend a desktop tool, like Kompozer, which would be much easier to work with? Web-based software definitely has its place, but I think that most people still prefer to use desktop based word processors, spreadsheets and IDEs.

  21. wintelkiller wrote on :

    interesting, but thias alpha release make no sence to work with …

  22. Harald wrote on :

    Talking about standards, I can only guess from context that the dates on this site are in US-type Month.Day.Year order, although I find those usually written with a slash.

    The initial release of this page in ISO notation ( would then be 2009-12-02, I guess.

  23. jeeremie wrote on :

    Really impressive app. However, I cannot type curly braces.

    Auto-completion would be a nice feature a tabbed interface as well.

  24. manielse wrote on :

    Interesting, I’ve been working on a project named Bespin for several months now (A video podcast series around news in the cloud). Though it doesn’t surprise me that someone else thought the name fit in the cloud computing era, the two products shouldn’t be confused with one another.

  25. Ezra wrote on :

    I just played around with Bespin and I am incredibly impressed. The app is extremely fluid and easy to use.

    One thing that concerns me is the background and font colors. The dark background and dark font don’t contrast well and make it hard to look at for long periods of time. The user should have the ability to change this.

    Anyway, GREAT work! I am always impressed by Mozilla.

  26. Michael Puckett wrote on :

    Outstanding concept and initial implementation! Will be tracking this product with great interest.

    Is it a reasonable assumption that processing tab key events is high on the task list? i.e. hitting the tab key with an active selection results in said selection being deleted, straying from typical editor behavior (indent in/out).

    Off to play with it some more …


  27. Simon wrote on :

    It’s possible to use with other services? For example can I host it on my own as an add-on to other services?

  28. Andrew Meredith wrote on :

    Sounds like a tall order, an excellent plan and good start.

    Good luck folks 🙂

    Andy M

  29. Anson Ho wrote on :

    I hope can translate the chinese version in this tool.Can we do that?

  30. sylt wrote on :

    This is really promising work! It will be very interesting to see where you can take this.

    Re: the collaboration demo…

    In the real world I personally find “instant update” collaboration to be at bit award and scary. I like to have control over whats going on with my stuff. I realize that this is mostly playing around with ideas and concepts on your part but I’d like to share my view on this subject anyway. 🙂

    The direction I would advocate for this type of editing is an editing mode where the user is made aware of changes made by others in real time but still retain the control over when, and what parts of the code, that is updated locally. As an example of how that could be implemented is to indicate graphically what lines/code snippets/functions have been modified, by whom, and to what extent. Then let the user at any time apply the most recent version (or from specific user or whatever) of a line/code snippet/function or update the entire file etc.

    I guess there are many different opinions on how things like this are to be done but this is my 2c 🙂

    Hope you keep playing around with this technology! It looks fairly awesome so far!

  31. monmonja wrote on :

    This is super cool, just want to request, can you put a shortcut to go to the command from the editor and vice versa coz using the mouse to switch between them slows down productivity (at least for some)

  32. Tom Potts wrote on :

    I’ve been playing with a browser JS editor in my spare time so I could do almost everything I need online.
    Thanks for this
    all the world needs now is JS2 and a svg editor in svg and the desktop is dead.

  33. Julien Couvreur wrote on :

    Very interesting.
    I especially like the re-use of Ubiquity APIs to create linguistic commands.

    On the other hand, creating a graphic toolkit on top of canvas seems really wacky. That seems really un-webish. Also, it is hard to imagine the performance would be better than a tweaked and native rendering engine. That said, I understand that the browser APIs may not be the best to do what you need…

  34. modeless wrote on :

    @ben I really hope you mean that about migrating the features into Gecko eventually. I’m imagining a future where HTTP and HTML are merely wrappers for opaque Javascript code that implements a complete UI toolkit in a fullscreen Canvas element, and I don’t like it one bit. That future is no better than one where Flash or some other proprietary thing replaces HTML. I think it’s absolutely the wrong direction for the Web, even though otherwise I love the ideas behind the Bespin project.

  35. Timothy Farrell wrote on :

    These guys need to get out a little more. They have good goals, but each of their goals (perhaps with the exception of the heatmap) is already accomplished by another project. Amy-editor has collaborative editing, codemirror can edit large files faster than Bespin can. But my point is that none of these use the canvas tag. Canvas is bad because I can’t make a quick edit on my iPhone. Since when have you ever seen an Internet Cafe that had Firefox on them? They’re defeating their own goals by doing it in a canvas tag. Instead they should resurrect Amy-editor and port the rendering engine from Codemirror over to it.

  36. Kevin Sawicki wrote on :

    What is the plan for integration with a source code management system?

  37. MyName wrote on :

    Thanks guys but i won’t try this. I’m too bored to register. Why should i register just to try the demo anyway? 🙁 Cm’on…

  38. Sylvain Pasche wrote on :

    About Linux performance: with 3.0 it’s indeed very slow (as in unusable) on my system (Ubuntu 8.10, nvidia, no visual effects). But on 3.1 (and 3.2) the speed is quite acceptable.

  39. Ben Pollak wrote on :


  40. John Taber wrote on :

    Hey Ben,
    nice – I’ll show this off in my IT Project Mgmt class re collaboration. I really like the dashboard – maybe there can be language specific versions ie in Rails a column for controllers, for models, for stylesheets, …. that would be really productive when jumping between files.

  41. DrorHarari wrote on :

    Looks very promising though still rough. The main issues I saw was blurry text (I used Chrome and 120dpi screen – one could expect trouble dealing with this when using a canvas) and the ease with which I could loose my edit by pressing the Backspace key (the browser went to the previous page loosing my changes) – the changes must remain!

    One other thing, in the collaboration mode, it would be nicer if every user had a private view in addition to a shared, common view. This way both can work on the same code but also each can look aside without disorienting the other.

    I’ll try to post bugs on the above…

  42. kib2 wrote on :

    Nice project: congrats !

    I’ve noticed something anoying: placing the cursor with the mouse worked until I was on the last one in my file.

  43. Eric the .5b wrote on :

    The limited cursor-navigation and it not retaining indentation makes it less than useful for me in practice, but it’s a nice start.

  44. Ben Galbraith wrote on :

    @Robert Kaiser: Unfortunately, the state of copy/paste integration with browsers via API is not good. We’ll get this fixed so it works seamlessly.

  45. Ben Galbraith wrote on :

    @Kas Thomas: RE slowness. What’s your platform? On Mac/Windows Firefox/WebKit Nightly it runs like a bat out of a hell on most platforms we’ve tested.

    Chrome runs quite slow–we’re not quite sure yet but we have some untested theories.

    We’ve heard reports that Linux browsers run slowly; we need to look into that.

    RE built-in help. We *completely* agree with everyone you’ve said in this regard. Coming soon. 🙂

  46. Ben Galbraith wrote on :

    @Dominic Burns: Definitely agree. We have lots of ideas for how local file system interactions might work; would love to hear yours.

    RE: “Mash up”. I take a small measure of comfort in knowing people far smarter than I who use the term too. 😉

  47. Shaun Santa Cruz wrote on :

    Nice! This project has alot of potential.

  48. Ben Galbraith wrote on :

    @Tomer: Great ideas, thanks very much for sharing. Please join our Google Group (link in the post above) and file issues on Bugzilla as your time permits. Your input is vital as we figure out where to take Bespin next.

  49. Ben Galbraith wrote on :

    @Marta: Unfortunately, we haven’t spent any time supporting non-US locales for this release. This was an oversight that we’ll correct as soon as we can. If you have the time, please file a bug at so we can be aware of as many issues as possible.

  50. Ben Galbraith wrote on :

    @Martin Tyler: We want to support both the use case of a team working on a public Bespin instance (a la GitHub or SourceForce) or that of a team installing it behind the firewall for private use.

  51. ben wrote on :

    @Dao: not necessarily. We’ve experimented with putting text / markup inside the canvas element itself to make it screen-reader friendly. This includes having the text in the editor in there, having a component hierarchy represented, etc.

    In fact, when you think about it, you can actually create *more accessible* applications that way as you could have a “logical” hierarchy representing semantic structure, as opposed to the somewhat messy real-world HTML hierarchies which inevitably include hacks to achieve desired visual effects in otherwise somewhat semantic markup.

  52. ben wrote on :

    @modeless: We really wanted Bespin to be cross-browser compatible, so extending Firefox to do new things wasn’t really as interesting to use as getting existing, widely-used browsers to do new things.

    That way, we can still look at features in canvas/JS that make sense to migrate into Gecko or WebKit but whilst users can use them in the meantime.

  53. Rodney Courtney wrote on :

    vi like support would be great. If the syntax would be the same as far as possible then it would be even better. So vi fans like me could immediately use this new tool too.

  54. Neil Stansbury wrote on :

    Very nice indeed. Though I do agree with the comments on canvas vs XBL elements, but I guess Nvu is there for that.

    I have to say my life has never been the same since we started using a modified version of Codepress inside a modified version of TiddlyWiki. It was in our XULUStudio app but I never got around to making it releasable quality.

    Each JS object & function becames a wiki snippet, and all the code is stored as a single wiki document. Dead easy for extracting, searching and building out of the wiki itself.

    Good luck though.

  55. Peter wrote on :

    This is really a fantastic job guys.

    It’s great to see a project that says, “This is what the cutting edge browsers can do. There are browsers that can’t handle some stuff. Lets write for the real innovators.” I remember listening to Audible Ajax when you were talking to the Aptana guys about Jaxer and they were saying ‘Mozilla has all this great stuff that no one uses.’ It’s excellent that the developers can dictate the limits of technology, rather than technology dictating limits on developers.

  56. Paul Colton wrote on :

    This is an incredible start to a very cool project. We’re pretty excited here at Aptana about Bespin, as it fits right into our Aptana Cloud service now and has a lot of potential with regards to where Aptana Studio could go in the future. Great work guys!

  57. ejunker wrote on :

    Bespin is cool but Amy Editor has more features.

  58. Robert Kaiser wrote on :

    This sounds like a cool experiment, but I see a potential problem in applying the general look & feel of the rest of the apps one is usually working with to a cavas-based environment.
    Apart from the usual problem as not being easily able to look at the “source” of your environment with a simple “vie source” context menu (which is usually the nice thing about “the web”), this seems to be a real killer to my fun with such a tool.

    My personal example for this is that I’m used to rely heavily on the copy-on-mouse-select / paste-with-middle-click-anywhere behavior of the X windowing system, across all kinds of applications on my computers. This work easily with normal websites and web apps, but I guess it won’t just work with such a canvas-based environment, and therefore render this editor unusable for me.

  59. Kas Thomas wrote on :

    Nice first go. Please continue.

    1. I grow oh-so-weary of the tired “adobe charcoal” look. Can I reskin this thing? I cannot and will not look at a black editor all day while writing code. Show-stopper.

    2. Sluggish. Unusably slow.

    3. The UI may be intuitive to someone else, but to me it’s cryptic. That’s okay but at least give me some contextual help right in the app. (please don’t send me to another page, I want huge mofo tooltips or something)
    Bottom-line: in-context help. No manuals. Should be self-documenting.

    4. Some Greasemonkey code-completion or typeahead would be nice.

    5. Built-in help with social APIs (Twitter) and REST APIs (OpenCalais) would be nice nice nice. Killer nice.

    This is an awesome first cut. Please continue full speed ahead. Congrats!

  60. SlugO wrote on :

    This seems quite exciting! Just a few problems I’ve noticed though: I can’t enter non-English letters like ä or ö and it doesn’t seem to work with Firefox 3.1 beta 2 in Arch Linux. Won’t open any projects.

  61. kourge wrote on :

    I’m happy to see that prototype.js was used; a nice choice since Bespin is by no means a thin client.

  62. Dominic Burns wrote on :

    Something needs to be done about a local interface and a local repository – it’s all well and good having all the things mentioned in the vimeo, but if the network’s down ‘imagine not being able to access your code _anywhere_’. Also, please, please don’t use the phrase ‘mash-up’ – it makes you sound like the sort of person who uses the phrase ‘mash-up’.

  63. Cícero Feijó wrote on :


  64. Tomer wrote on :

    This is great effort, but some features are missing –

    A. Publishing – Am I the only one finds it strange that the editor don’t allow me to export the files to my own disk or server? I would be even happier if a server-side PHP script wrapper will allow publishing to remote SSH servers.

    B. Caching – The editor is not too fast if browsing from remote. You can use HTML5 specs to make it faster and more powerful.

    C. Characters – Typing non-latin letters doesn’t appear on the screen and neither written to the file. This make it impossible to use the editor for other languages.

    D. Client-Side Customizations – by using the power of client-side scripting such as Greasemonkey for Firefox, we can allow anyone to customize the editing environment for their own.

    E. Version Control – I am missing some kind of basic version control built into the IDE. This can be made by saving backups for every file, and allowing side-by-side comparisons between each and every saved version.

    F. Built For Standards – I would really appreciate adding standards to the view. For example, new HTML documents should always include DOCTYPE, and the editor should warn if IMG tag has missing ALT attribute.

  65. Sebastian Macias wrote on :

    Very nice guys,


  66. Marta wrote on :

    I can’t get curly braces to appear using my portuguese keyboard… any suggestions?


  67. Martin Tyler wrote on :

    Interesting project. I am curious about the use cases for this. For example, if a team was developing a website and wanted to use this, do they install it on their dev server and edit the website directly? What about issues with the location of the files, eg you want your website to be accessed from the root of the webserver? Then what do they do when they want to move the site to production servers? Also any thoughts on source control systems etc?

    I know this is an early demo, so I am asking about the vision rather than criticising its current state.

  68. Gen Kanai wrote on :

    Some press coverage:

    Mozilla Labs lands on Bespin

    Bespin: Could Your Favorite HTML Editor Live in the Cloud?

    Introducing Bespin, from Mozilla Labs

    What Mozilla’s Bespin Bespeaks

    Mozilla presenta Bespin, piattaforma per sviluppo collaborativo

    Mozilla asks developers to take Bespin for a spin

    Mozilla Ships Bespin Web-Based Code Editor

    Mozilla Bespin tries taking coding to the cloud

    Mozilla Bespin Is a Killer Web-Based Text Editor

    Bespin Web-based code editor effort launched

    Mozilla Labs Introduces Bespin For Collaborative Coding

    Mozilla Labs Releases Bespin Web-Based Code Editor

    Mozilla Labs preview Bespin online code editor

    Mozilla Introduces “Bespin” for Programming in the Cloud

  69. Dao wrote on :

    Isn’t the extensive use of canvas bad for accessibility?

  70. modeless wrote on :

    This is an great project, but I am disappointed that Canvas is used for everything. It’s obviously true that you can prototype some neat features that way, but I think the ideal outcome of this project would be to roll all these great features into Gecko and so they can be used by every Web developer.

    Instead of developing Bespin and Thunderhead in Canvas and completely ignoring Gecko, wouldn’t it be cool to make a branch of Gecko and add all the features Bespin and Thunderhead need, with an eye toward merging to trunk and eventual inclusion in W3C standards?

    For an easy example, truncating items with an ellipsis in the middle is just begging to be added to CSS as text-overflow:-moz-ellipsis-center or something like that. For performance on large files, the bottlenecks that cause slowdowns could be identified and addressed. In general the crazy layout stuff going on in the Thunderhead UI toolkit is not something that would be out of scope for maybe CSS 4 or 5, and prototyping it in Gecko now would give Mozilla a huge head start and more influence in the future standards process.

    The WebKit guys have been busy adding crazy new rendering engine features like CSS animation and (3D!) transforms, and I’d love to see Mozilla implementing pie-in-the-sky new Gecko features just like that. Doing it while simultaneously working on real projects that actually require the new features is the absolute best way to do it.

  71. ben wrote on :

    Thanks to everyone for the feedback! We really appreciate it.

  72. ben wrote on :

    @Ferodynamics: Thanks, I think?

  73. ben wrote on :

    @Jim Battle: Actually, high-quality vi and emacs modes are on our list; keep watching.

  74. ben wrote on :

    @Dan: Sorry about that, we’ll see if we can reproduce the flakiness on our end. If you find a definitive way to reproduce, would be very helpful if you could file a ticket.

  75. ben wrote on :

    @Nicholas: We would love to spin out the core editor as a component folks can use for applications like CMS. Look for something to enable that in future versions.

  76. Andreas wrote on :

    hi, bespin looks great, please add support for non english keyboard layouts 😉

  77. Mario wrote on :

    Cool stuff, guys!
    I’m really looking forward to have vim features in that editor.
    Just downloaded the source, now trying to run Bespin on my server. Please always think about this use case. Many developers will be interested in not running the editor in the “cloud” but on their own servers.

  78. Jann Aho wrote on :

    This seems awesome!

  79. Jakub Steiner wrote on :

    The project is too cool to have a save ‘function’. Drop the save button!

  80. Wayne wrote on :

    Are there plans to allow for emacs & vi style key bindings? I must admit, that I’d probably not use it without those bindings.

    Otherwise, it looks fantastic and what’s there works really well. Nice one.

  81. AjitK wrote on :

    Its marvelous!

    I had been yearning for this for a while. I am sure it can be integrated with Firebug to help increase developer productivity!

  82. Tom wrote on :

    This is a really awesome idea. This first version is a real achievement. It has a lot of the exact same issues the Heroku editor does (like cursor movement and selections are screwy, key bindings misbehave, etc.), but it feels pretty quick. There is still sluggishness in cursor movement, which is a major issue, but I’m sure there are optimizations that can be done.

    You should get in touch with the Heroku guys, they built a full browser code editor, and I believe they did it all from scratch with standard HTML/CSS/Javascript, including writing all the nasty stuff (cursor, text rendering, selection handling, etc.) themselves as well. Of course they built theirs in late 2007, so I’m sure you’ve got better browser capabilities to work with now (like HTML5).

    Plus they figured out the really key issue: how do you edit the files and see your changes reflected in a running app somewhere? After all, you can’t edit in a vacuum.

    Your collaboration ideas are exactly the right way to go. Nice work!

  83. PJ Brunet wrote on :

    I recommend you talk with the guys, they have similar goals I think

  84. Ferodynamics wrote on :

    This is the coolest IDE since Turbo Pascal 7.0 😉

  85. Jim Battle wrote on :

    Call me back when it has vim emulation mode.

    I don’t care how great an editor is; I have vi neurons that were built 25 years ago, and rewiring that tangle could sap any theoretical efficiencies of the new editor.

    (I’m half joking, but half not)

  86. Mark Holton wrote on :

    Congrats, Ben and Dion and Bespin team, this is really exciting to see! Looking forward to seeing this grow.

  87. balanceshift wrote on :

    I’m impressed, and excited for this. Guessing the title is referring to the planet Bespin and Cloud City? I like it!

  88. Dan wrote on :

    I’m having a bit of trouble… sometimes logging in doesn’t work, and when it does, the Dashboard just kicks me back to the home page. Happens in both Firefox and Google Chrome.

  89. Nicholas Orr wrote on :

    Is there going to be a simple textarea control for CMS content.

    WYSIWYG editors suck, I just want colour coded html editor that preserves indents, tags,entities etc

  90. Jay Goldman wrote on :

    Congrats guys!

    I had no idea that it was releasing today when I saw the demo 🙂 Really exciting step forward and I can’t wait to see some of the other stuff you mentioned was coming down the pipe!

  91. Sugendran Ganess wrote on :

    Holy crap this is awesome.