IBM suspending work on Maqetta

5,614 views
Skip to first unread message

Jon Ferraiolo

unread,
May 10, 2013, 1:07:48 PM5/10/13
to maqett...@googlegroups.com
It is with great regret that IBM is suspending its work on the Maqetta open source project.

For the time being, IBM will continue to maintain maqetta.org, administer the open source project at GitHub, and administer various other supplemental things associated with Maqetta (e.g., YouTube channel, Twitter account, etc), on an as-needed basis.

We hope that other open source developers will appear to take over leadership on Maqetta. Over the next month or two at least, the existing committers are available to help people who might want to take over.

The Maqetta open source team is grateful for the partnership we have had with the community, and we are glad that many users and companies have found value in the product. The team is also proud of how we pushed the boundaries of browser-based tooling, and hope that other developers will take it from here.

Thanks to everyone who has contributed to the Maqetta effort, either as users or developers.

Jon Ferraiolo, IBM

David Curry

unread,
May 10, 2013, 4:50:55 PM5/10/13
to maqett...@googlegroups.com
A little explanation might help here, wouldn't you say? The web site
doesn't even mention this action. And the term "suspend" is odd,
implying that development could be resume in the hopefully-near future.
Even more odd is that, prior to this announcement, transfer of control
to Apache, SourceForge or another FOSS supporter wasn't worked out already.

I'd like to know why, though:
No future for this type of product? Nah, HTML5 is still all bright 'n
shiny.

Another project that has better prospects, perhaps commercial rather
than FOSS? Hmmm.

Maqetta itself going commercial, being taken out of the FOSS community
now that we've helped make it a better product? Meh, not really IBM's
style, and really kind of small taters anyway for a move like that.

IBM is completely giving up on software development? splch, snarch,
yeah, right.

Ahhhh, Maqetta needs some time to evaluate the architecture for a
potential really cool refactoring/rearchitecting to take it to the next
level, and the team doesn't want to be hassled with the kind of
support/questions a live project would get, needs to quiet time - that's
what I think.

Well, at least these releases don't need an IBM server to run, so what
ya got, ya got and can use, and let the community 'own' it if it is
really that good. And if we never hear of Maqetta again, thanks Bmers
for all your help...it was a fun ride!

Dave
> --
> You received this message because you are subscribed to the Google
> Groups "Maqetta Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to maqetta-user...@googlegroups.com.
> To post to this group, send email to maqett...@googlegroups.com.
> Visit this group at http://groups.google.com/group/maqetta-users?hl=en-US.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

David Curry

unread,
May 10, 2013, 4:52:33 PM5/10/13
to maqett...@googlegroups.com
And why didn't anyone else post a whine/gripe/condolence??? It's been
almost 4 hours, for gork's sake, a frelling eternity!

On 05/10/2013 10:07 AM, Jon Ferraiolo wrote:

Javier Pedemonte

unread,
May 10, 2013, 5:23:35 PM5/10/13
to maqett...@googlegroups.com
The complete source code is available on GitHub and will not be taken down.  Anyone is welcome to fork the code and start their own Maqetta derivative, if they so choose, and generate their own builds.

maqetta.org is run by IBM and as Jon stated, we will continue to do so, for now.  Of course, anyone else is welcome to deploy their own Maqetta instance.

We'll update the main site as well...


Javier
- maqetta team

Scott Elcomb

unread,
May 10, 2013, 5:47:53 PM5/10/13
to maqett...@googlegroups.com
On Fri, May 10, 2013 at 4:52 PM, David Curry <curr...@mchsi.com> wrote:
> And why didn't anyone else post a whine/gripe/condolence??? It's been
> almost 4 hours, for gork's sake, a frelling eternity!

After being active with FOSS projects over the last decade, I've seen
tonnes of projects come and go. Hell, I've got projects that've
barely been touched in years; they're suspended but I've not forgotten
and still pick away at them periodically. I for one won't complain -
I appreciate the work that IBM's put into the project and respect
their decision to step away from it.

In some ways I find the announcement disappointing (it was rather
sudden) but in other ways I find it promising - I'd welcome a rewrite
for example.

My own career in web development started because I wanted to build
something similar in nature to Maqetta. Without getting into the long
of it, my most recent attempt (proof-of-concept only) came as a result
of trying to show the difference between Maqetta and a custom built
alternative. Where Maqetta is roughly 100M to download, my prototype
is 21K (in its SPA form)*. Obviously, Maqetta is far more capable and
applicable in many more situations. But it's also incredibly complex;
I've been hacking away at it for almost a year and still have only a
limited understanding.

I do hope to see development continue, but it's still a useful project
regardless.

Best,
- Scott.

* <https://github.com/psema4/Atomic-VE>
Live SPA demo [use Chromium & devtools] <http://psema4.github.io/Atomic-VE/>

Some metrics & thoughts re: comparison to Maqetta; these are provided
simply for reference and to back my claim of interest regarding a
Maqetta re-write. I do not - in any way - recommend this as a
replacement to Maqetta! (Please also note that metrics will be
slightly off-actual due to minor updates after these notes were
taken.)

Stats:
• index.html - 1.6K, 46L
• app.css - 3.5K, 188L
• app.js /min - 42K/16K, 781L
• total: 47.1K/21.1K, 1015 Lines
• Total time to produce: ~40h

Pros:
• Zero dependencies, no external libraries of any kind are required
• Extremely lightweight; complete application (html+js+css) is 21K
(with JS minified)
o For comparison this POC is 11K smaller than the current minified
jQuery distribution (v1.8.2 @ 32K) and almost 5000 times smaller than
the current Maqetta release (v7.0.1 @ 99.6M)
• Easy to add new widgets and widget-specific attributes
• Basic inline editors for primary attributes
• No iframes
• Preview functionality
• Total line count of all files (1015 lines) includes blank
(whitespace) lines as well as comments
• Atomic capability (single document SPA [Single Page Application], no
external files whatsoever)
• Does not require a server

Cons:
• Features not implemented:
o drag-select multiple widgets
o snap-to-grid, snap-to-widget
o re-parenting widgets
o widget resize handles
o attribute-specific input widgets in Attributes window
o pass widget & attribute names to inline editors (eg editor for
img widget edits src property)
o selection indicator in inline editors (for radio/checkboxes, like maqetta)
o inline editor for complex list-based widgets is incomplete
(radio- and check-groups)
• Shortcuts taken:
o repurposed widget drag-and-drop for dialog positioning (window
selection can sometimes cause the window to reposition)
o menus are simple select elements
o two methods for widget identification (“hints”) - one for basic
html widgets (inferred from tagName) and another for complex widgets.
All widgets should use the second method (where widget and property
hints are stored as data-* attributes in DOM nodes)
o No cross-browser support, this POC has only been tested in Chrome

--
Scott Elcomb
@psema4 on Twitter / Identi.ca / Github & more

Atomic OS: Self Contained Microsystems
http://code.google.com/p/atomos/

Member of the Pirate Party of Canada
http://www.pirateparty.ca/

cjs

unread,
May 10, 2013, 6:02:24 PM5/10/13
to maqett...@googlegroups.com
Well, I'm disappointed.
IBM has now killed 2 Projects that I've been trying to use.
Maqetta and the EGL Development tooling Eclipse project.

I guess IBM has decided there is nothing in it for them.

Rick Lacy

unread,
May 10, 2013, 6:14:17 PM5/10/13
to maqett...@googlegroups.com
Sad news indeed.  This project made significant progress since its inception and really pushed the boundaries of HTML WYSIWYG.  I was looking forward to seeing how it progressed in the future, but now that future is uncertain.

I suppose we shouldn't be too surprised.  IBM has made significant contributions to FOSS, but at the end of the day projects must be prioritized and revenue must be made.  Things change over time.

This isn't the first time I've been disappointed and won't be the last.  It's a regrettable loss of innocence when you realize that it isn't always the best or coolest idea that 'wins'.   

Take dojoserverfaces.  


This sounded like an amazing idea.  Combine the power of Dojo with the ease of use of JSF.  IBM included beta tooling for it in RAD 8.0, had articles on developerworks, and put up a site for it.  However, the lead was apparently relocated within IBM and the project died.  First the tooling was removed from RAD 8.5.  Then one day the site for it disappeared.  Now all that's left is a github repo where the freshest commit is two years old.  I'm thankful that at least the maqetta team was courteous enough to let us know something was going on.

I hope someone takes over development, but that is far from certain.  I wish I had the time to devote to it... I'll probably at least look at the source but there's never enough time in the day.

More alarming to me than just the suspension of Maqetta is what this says about dojo.  Maqetta has long been held up as a shining example of what it is possible to do in dojo, and it was also a nice public example of IBM's commitment to dojo.  

To the point made previously, no, obviously IBM is not getting out of the software game entirely.  But maybe they're sick of playing the JavaScript framework user interface game.  I've been a little alarmed that RAD's support for AMD syntax is still spotty at best and that instead of improving it focus has been on things like support for jquery mobile.  Come to think of it, I'm a little alarmed by AMD in general since I loathe the syntax so much...

The loss of maqetta is just one less reason to recommend that people consider dojo for their next project.  For that matter, why the heck aren't the dojo developers and / or sitepen people eager to take this over?  Are they all too busy fighting over 2.0 to see what they'll lose here?

There are also solutions in RAD that are in a similar vein to maqetta - like the Rich Page Editor - so maybe IBM decided to focus on these to be better integrated with their existing toolchain.

Who knows.  Of course this is all simply speculation.  As has been said, FOSS projects come and go all the time.  Just a warning that buying into the latest and greatest ideas is sometimes dangerous and leads to disappointment.

Big thanks to the maqetta team for all their excellent work and for showing us what's possible.  Your efforts were an inspiration and will be missed.

On Friday, May 10, 2013 12:07:48 PM UTC-5, Jon Ferraiolo wrote:

David Curry

unread,
May 10, 2013, 6:19:19 PM5/10/13
to maqett...@googlegroups.com
Scott,
I, too, respect IBM's decision, and am not questioning IBM's right to
make it. I really do wonder, though, if "you liked Maqetta, but you'll
love ???" is in our future.

I am, however, pleased to see some chatter about that decision. Think
about it - if you stop production on something, and no one complains,
then maybe the product never was worth the effort and expense. Perhaps
enough dammit's will convince IBM to resume-from-suspend. :D

Oh, I know - management brought in a Watts Humphrey acolyte, wants the
team to use PSP, and the team has revolted! Oh, dear!

Best luck with your project, I will be checking it out.
Dave
> � index.html - 1.6K, 46L
> � app.css - 3.5K, 188L
> � app.js /min - 42K/16K, 781L
> � total: 47.1K/21.1K, 1015 Lines
> � Total time to produce: ~40h
>
> Pros:
> � Zero dependencies, no external libraries of any kind are required
> � Extremely lightweight; complete application (html+js+css) is 21K
> (with JS minified)
> o For comparison this POC is 11K smaller than the current minified
> jQuery distribution (v1.8.2 @ 32K) and almost 5000 times smaller than
> the current Maqetta release (v7.0.1 @ 99.6M)
> � Easy to add new widgets and widget-specific attributes
> � Basic inline editors for primary attributes
> � No iframes
> � Preview functionality
> � Total line count of all files (1015 lines) includes blank
> (whitespace) lines as well as comments
> � Atomic capability (single document SPA [Single Page Application], no
> external files whatsoever)
> � Does not require a server
>
> Cons:
> � Features not implemented:
> o drag-select multiple widgets
> o snap-to-grid, snap-to-widget
> o re-parenting widgets
> o widget resize handles
> o attribute-specific input widgets in Attributes window
> o pass widget & attribute names to inline editors (eg editor for
> img widget edits src property)
> o selection indicator in inline editors (for radio/checkboxes, like maqetta)
> o inline editor for complex list-based widgets is incomplete
> (radio- and check-groups)
> � Shortcuts taken:
> o repurposed widget drag-and-drop for dialog positioning (window
> selection can sometimes cause the window to reposition)
> o menus are simple select elements
> o two methods for widget identification (�hints�) - one for basic

Wagner Marques

unread,
May 11, 2013, 11:03:21 AM5/11/13
to maqett...@googlegroups.com


I can not believe it!!!

Seams to me  that the market  adopt and invest not always in the best TI option but in the most visible option.

I have not compentence to help with coding for the project but I am expecting that the term "suspend" means that the IBM will resume the works on maquetta.

Jon Ferraiolo

unread,
May 11, 2013, 1:26:43 PM5/11/13
to maqett...@googlegroups.com
 I am expecting that the term "suspend" means that the IBM will resume the works on maquetta.

What I meant by "suspend" is that IBM has pulled its engineers from working on Maqetta, and that IBM might or might not choose to resume work on Maqetta at a future time. But in terms of probabilities of whether IBM will resume work, another sentence in the email said: "We hope that other open source developers will appear to take over leadership on Maqetta."

Jon Ferraiolo, IBM


--

David Curry

unread,
May 11, 2013, 1:55:18 PM5/11/13
to maqett...@googlegroups.com
Jon,
Please accept and pass the word to said engineers what I am sure is a community-wide appreciation for all the work that went into Maqetta, as well as for the continued support.  IOW, THANKS!!!

Wagner Marques

unread,
May 11, 2013, 6:13:10 PM5/11/13
to maqett...@googlegroups.com
I agree, with you David!!!


2013/5/11 David Curry <curr...@mchsi.com>

--
You received this message because you are subscribed to a topic in the Google Groups "Maqetta Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/maqetta-users/dLGzdERrkG8/unsubscribe?hl=en-US.
To unsubscribe from this group and all its topics, send an email to maqetta-user...@googlegroups.com.

To post to this group, send email to maqett...@googlegroups.com.
Visit this group at http://groups.google.com/group/maqetta-users?hl=en-US.
For more options, visit https://groups.google.com/groups/opt_out.
 
 



--
Wagner França Marques
54731643 (vivo)
70768470 (claro)

mc007ibi

unread,
May 16, 2013, 7:54:50 AM5/16/13
to maqett...@googlegroups.com
Hi,
we're developing a Dojo based mobile app studio which works already quite ok. We do have bigger plans with Maqetta and we will implement it as GUI designer for our suite in this year. The first cycle is to remove all Java and Eclipse server side code and deps. After this, we should be able to deploy it more easy and use Node.js or others for server side. Please stay in touch here xapp-studio.com.

Thank you, its really amazing work and we admire the entire design and implementation done in Maqetta. We hope we can build a complete web based IDE with Maqetta, open source of course. 
guenter

Javier Pedemonte

unread,
May 16, 2013, 10:31:22 AM5/16/13
to maqett...@googlegroups.com
The first cycle is to remove all Java and Eclipse server side code and deps. After this, we should be able to deploy it more easy and use Node.js or others for server side.

Sounds interesting.  Please note that since M7, Maqetta has been built on top of Eclipse Orion and makes extensive use of it, particularly on the server.  And the Orion server is built on Eclipse Equinox (Java).

For the Orion project, there "reference" implementation is this Java Equinox Server.  However, they are also working on a Node.js based server, which should be of interest to you.  It implements some but not all of the functionality of the Java server -- in particular, it is currently a "single user" server and doesn't provide the user management capabilities of the Java server.  So keep track of what they are doing with their Node.js server....

Javier
- maqetta team 

mc007ibi

unread,
May 16, 2013, 3:18:10 PM5/16/13
to maqett...@googlegroups.com
Hi,
thank you, yes. I will get in touch as soon we're in the phase of 'cleaning'. For my taste, Maqetta shouldn't need any heavy server in the single user mode. It should be a plug-able JS only framework/app first and then support simple ways to distribute it to Mac-App store and friends for instance and make integration easy, and that's our first objective. But hey, for true,  Node.js isn't really the answer either, I agree. I rather thought about a combination of it with Node.JS or/and Cordova for desktop because most of web developers love freedom and fast offline apps. That could give it the right drive it needs,...

well, see you soon on the other side :-)
guenter

Piotr Piwowarczyk

unread,
May 19, 2013, 9:19:51 AM5/19/13
to maqett...@googlegroups.com

Oh dear, I need to migrate all my projects now. As I know large companies Maqettab won't stay online for a long time.

Piotr

--

mc007ibi

unread,
May 19, 2013, 10:39:22 AM5/19/13
to maqett...@googlegroups.com, mega...@googlemail.com
hi,
if you need proper hosting for Maqetta (free), let me know. We've got plenty of servers.
g

Javier Pedemonte

unread,
May 23, 2013, 5:34:49 PM5/23/13
to maqett...@googlegroups.com, mega...@googlemail.com
It's fairly simple to run Maqetta locally on your machine.  Download and run one of the maqetta.server.* scripts to start it up.  Then load http://localhost:50000/maqetta in your browser.

You can download your projects (one at a time) using one of two ways:
  1. From Maqetta, in the Files view in the lower left, select the Download Entire Project icon (4th from the right on 2nd row).
  2. Select Open > Orion Navigator to open Orion in a new tab.  Select one of your projects (click next to project name to highlight), then select Action > Export as zip.

Javier
- maqetta team

Gilbert

unread,
Jul 16, 2013, 3:44:52 PM7/16/13
to maqett...@googlegroups.com
Jon, First, I know it must be a little painful to discuss the root cause for IBM suspending further work on Maqetta. But will you?  Why is it doing this? Secondly, can you direct me to anyone at Maqetta, suspended or not, starting with yourself, who can assist me with coding some IF ELSE  javascript intelligence into a dojo Button I want to use in a maqetta prototype I'm building. Documentation on coding intelligence into dojo's onclick event seems to be non-existent.  Thirdly, beginners like me will miss you.

Gilbert Moore (an ex-IBMer), Blue Logic

Rob Weiss

unread,
Jul 16, 2013, 5:36:56 PM7/16/13
to maqett...@googlegroups.com, maqett...@googlegroups.com
Gilbert,
I can help out with your problem. I am typically on the dojo irc during the day; if you don't see me, j105rob, you can always email me directly at j105...@gmail.com

Sent from my iPhone
--
You received this message because you are subscribed to the Google Groups "Maqetta Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maqetta-user...@googlegroups.com.
To post to this group, send email to maqett...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages