Future of Prototype.js

1,464 views
Skip to first unread message

buda

unread,
Sep 23, 2011, 7:34:00 AM9/23/11
to Prototype & script.aculo.us
It has long heard nothing about plans for the future.
Browsers are evolving, and many library functions are duplicate the
functions of JavaScript.
Already implemented in all browsers support ECMAScript 5, but the
library is not reflected.
I would like to be able to not pull in hundreds of kilobytes to
support older browsers and have only the functionality you need - as
it is implemented in other libraries like JQuery.

What is the future of Prototype.js?
Is it live or dead? Should I start to learn JQuery?

T.J. Crowder

unread,
Sep 23, 2011, 12:19:54 PM9/23/11
to Prototype & script.aculo.us
Hi,

This was _JUST_ gone into, in depth:
http://groups.google.com/group/prototype-scriptaculous/browse_thread/thread/7d434e109c23c306

I think this is a fair-ish (but incomplete) summary:

1. Prototype is a spare-time activity for a very small number of
people. In fact, at the moment I think it's mostly just Andrew, and he
has other things he does (like, you know, hold down a full-time job).
Just recently the project largely lost another major contributor to
his job and family commitments.

2. Andrew has no plans to stop working on Prototype on any time soon.

3. Prototype work is done in fits and spurts, long periods of
inactivity followed by a period (couple of weeks) of feverish
activity.

4. jQuery has multiple corporate sponsors, not least Microsoft, and
there are people paid to work on it full time. Thus it's able to be
much more up-to-date and proactive than Prototype.

A quote from Andrew in the thread linked above:

On Aug 23, 1:07 am, Andrew Dupont <googlegro...@andrewdupont.net>
wrote:
> I will say, though, that if we're crowning winners and losers, then
> jQuery "won" a long time ago. It is certainly the _de facto_
> JavaScript library for web development. The good news is that the
> "losers" of the war aren't looking so bad; libraries like Prototype,
> MooTools, and Dojo still have loyal user bases, and I doubt they're
> going away.

My personal perspective, for what it's worth (e.g., possibly exactly
what you're paying for it, nothing :-) ) is based on some data points:

1. The last blog post on the Prototype blog was 10 months ago.

2. The library has had basically no activity since v1.7 was released
November 2010.

3. The most recent three releases were:
v1.7 - November 2010
v1.6.1 - September 2009
v1.6.0.3 - September 2008

Three releases, in total, including "dot" releases, in the last three
years. Compare with 16 releases (four major ones) of jQuery in that
time period (v1.3 through v1.6.4).

4. Absolute use and trends:

Absolute use: http://trends.builtwith.com/javascript
Prototype trend: http://trends.builtwith.com/javascript/Prototype
jQuery trend: http://trends.builtwith.com/javascript/JQuery

5. Questions tagged on StackOverflow:

jQuery: 114,842
Prototype: 2,152

...which could, of course, just mean that Prototype is so much better
it generates fewer questions, or that people using Prototype don't use
StackOverflow, or that people are mis-tagging JavaScript questions
"jquery" (I see that a fair bit), or some combination of those.

Does all this mean Prototype is dead? No, not a bit of it. But it has
a very small staff with other major demands on their time, and has no
funding. Andrew's quite clear that it's not dead, and also that it
will continue in much the way it has these last three years.

I made the business, not technical, decision years ago to use jQuery
rather than Prototype. In many ways I prefer Prototype, although there
are some good ideas in jQuery (also some phenomenally bad ones, such
as how overloaded the API is). I still pitch in and moderate this
mailing list, and still answer questions, but for me the business case
is: Which library amongst the large number out there is kept up-to-
date; jumps on testing new browser versions for compatibility; has a
large ecosystem of code I can use; has a large pool of talent I can
hire or contract; has well-maintained, frequently updated and
extended, reliable, and documented UI helpers (jQueryUI vs.
Script.aculo.us); and is likely to be around long-term without heroic
effort from a single individual, or a small set of individuals. So I
went with jQuery, despite preferring Prototype in many ways. [I also
looked at Dojo, ExtJS, (more recently) Closure, and a few others.]

It could have gone another way. People like me could have done more to
contribute to the project; leadership could have focussed on core
functionality, community-building, and developer (um) development
rather than side-issues; corporate sponsorship could have been courted
and perhaps ultimately found. But that didn't happen, and it didn't
happen because the Prototype community and leadership didn't make it
happen (perhaps corporate sponsorship wasn't desirable; fair 'nuff)
and because luck didn't go Prototype's way. (And don't think luck
isn't a big factor here.)

I have nothing but respect and admiration for Andrew and everyone else
who has made Prototype what it is. And there's absolutely no reason
not to use it on your websites if you test with your target browsers
and it does what you need it to do. Again, Andrew is clear he'll keep
going with it, and I'm sure he'd be very happy to have help from
anyone reading this.

So does Prototype have a future? Yes. What kind of future depends a
lot on the people reading this note. If you all, like me, don't have
time to contribute and Andrew has to largely work on his own, that's
one kind of future. If you can make time to help, get your companies
to let you help a bit during paid time, that sort of thing, then it
will have a different kind of future. I hope all these recent
questions about Prototype's future mark the beginning of a renaissance
for the library, a turning point of talent looking to help out. I wish
I could be part of it.

Best,
--
T.J. Crowder
Independent Software Engineer
tj / crowder software / com
www / crowder software / com

T.J. Crowder

unread,
Sep 23, 2011, 1:34:23 PM9/23/11
to Prototype & script.aculo.us
Hi,

I should have said: If you do decide to switch, beware that jQuery
only covers about 3/4ths of the ground Prototype covers. It does all
the DOM stuff, but doesn't do much of anything in terms of language
enhancement. So no equivalent to `Class`, for example, nor any of the
added methods on `Array`, `String`, or `Function`. (It does have a
subset of those on offer, just not as extensions to the built-ins --
for instance, `jQuery.trim` rather than `String#strip`, `jQuery.proxy`
rather than `Function#bind`). If you need a `Class`-like thing, I did
one you might look at:
http://blog.niftysnippets.org/2009/09/simple-efficient-supercalls-in.html

My main point being: jQuery is great, but its scope isn't as broad. In
some ways that may be what you want with this ECMAScript5 stuff coming
in -- or not.
--
T.J. Crowder
Independent Software Engineer
tj / crowder software / com
www / crowder software / com

On Sep 23, 5:19 pm, "T.J. Crowder" <t...@crowdersoftware.com> wrote:
> Hi,
>
> This was _JUST_ gone into, in depth:http://groups.google.com/group/prototype-scriptaculous/browse_thread/...

buda

unread,
Sep 23, 2011, 2:27:25 PM9/23/11
to Prototype & script.aculo.us
Thanks for the detailed post.
I myself have been using Prototype.js from the very beginning and its
ideology is close to me.
I'm anxious for the fate of the library - and therefore raised the
question
I would be pleased to help the project. but I do not know what and how
- the money that I earn in Ukraine, so small that they are not enough
for life to say nothing of assistance to the project:)
The only thing I could - to write what I feel right now must be
changed in the library.

For example:
- would love to be able to collect a library, pointing out the
necessary functionality
- be able to "drop" support for older browsers and crossbrowsing
(modern browser now have the same functionality)

Ready with pleasure to help develop the project - tell me how.

On 23 сен, 20:34, "T.J. Crowder" <t...@crowdersoftware.com> wrote:
> Hi,
>
> I should have said: If you do decide to switch, beware that jQuery
> only covers about 3/4ths of the ground Prototype covers. It does all
> the DOM stuff, but doesn't do much of anything in terms of language
> enhancement. So no equivalent to `Class`, for example, nor any of the
> added methods on `Array`, `String`, or `Function`. (It does have a
> subset of those on offer, just not as extensions to the built-ins --
> for instance, `jQuery.trim` rather than `String#strip`, `jQuery.proxy`
> rather than `Function#bind`). If you need a `Class`-like thing, I did
> one you might look at:http://blog.niftysnippets.org/2009/09/simple-efficient-supercalls-in....

buda

unread,
Sep 24, 2011, 2:22:22 PM9/24/11
to Prototype & script.aculo.us
I also nee optional parameter in 'observe' and 'on' methods to capture
events

On 23 сен, 21:27, buda <www...@pochta.ru> wrote:
> Thanks for the detailed post.
> I myself have been using Prototype.js from the very beginning and its
> ideology is close to me.
> I'm anxious for the fate of the library - and therefore raised the
> question
> I would be pleased to help the project. but I do not know what and how
> - the money that I earn in Ukraine, so small that they are not enough
> for life to say nothing of assistance to the project:)
> The only thing I could - to write what I feel right now must be
> changed in the library.
>
> For example:
> - would love to be able to collect a library, pointing out the
> necessary functionality
> - be able to "drop" support for older browsers and crossbrowsing
> (modern browser now have the same functionality)
>
> Ready with pleasure to help develop the project - tell me how.
>
> On 23 ÓÅÎ, 20:34, "T.J. Crowder" <t...@crowdersoftware.com> wrote:
>
>
>
>
>
>
>
> > Hi,
>
> > I should have said: If you do decide to switch, beware that jQuery
> > only covers about 3/4ths of the ground Prototype covers. It does all
> > the DOM stuff, but doesn't do much of anything in terms of language
> > enhancement. So no equivalent to `Class`, for example, nor any of the
> > added methods on `Array`, `String`, or `Function`. (It does have a
> > subset of those on offer, just not as extensions to the built-ins --
> > for instance, `jQuery.trim` rather than `String#strip`, `jQuery.proxy`
> > rather than `Function#bind`). If you need a `Class`-like thing, I did
> > one you might look at:http://blog.niftysnippets.org/2009/09/simple-efficient-supercalls-in....
>
> > My main point being: jQuery is great, but its scope isn't as broad. In
> > some ways that may be what you want with this ECMAScript5 stuff coming
> > in -- or not.
> > --
> > T.J. Crowder
> > Independent Software Engineer
> > tj / crowder software / com
> > www / crowder software / com
>
> > On Sep 23, 5:19špm, "T.J. Crowder" <t...@crowdersoftware.com> wrote:
>
> > > Hi,
>
> > > This was _JUST_ gone into, in depth:http://groups.google.com/group/prototype-scriptaculous/browse_thread/...
>
> > > I think this is a fair-ish (but incomplete) summary:
>
> > > 1. Prototype is a spare-time activity for a very small number of
> > > people. In fact, at the moment I think it's mostly just Andrew, and he
> > > has other things he does (like, you know, hold down a full-time job).
> > > Just recently the project largely lost another major contributor to
> > > his job and family commitments.
>
> > > 2. Andrew has no plans to stop working on Prototype on any time soon.
>
> > > 3. Prototype work is done in fits and spurts, long periods of
> > > inactivity followed by a period (couple of weeks) of feverish
> > > activity.
>
> > > 4. jQuery has multiple corporate sponsors, not least Microsoft, and
> > > there are people paid to work on it full time. Thus it's able to be
> > > much more up-to-date and proactive than Prototype.
>
> > > A quote from Andrew in the thread linked above:
>
> > > On Aug 23, 1:07šam, Andrew Dupont <googlegro...@andrewdupont.net>
> > > wrote:
>
> > > > I will say, though, that if we're crowning winners and losers, then
> > > > jQuery "won" a long time ago. It is certainly the _de facto_
> > > > JavaScript library for web development. The good news is that the
> > > > "losers" of the war aren't looking so bad; libraries like Prototype,
> > > > MooTools, and Dojo still have loyal user bases, and I doubt they're
> > > > going away.
>
> > > My personal perspective, for what it's worth (e.g., possibly exactly
> > > what you're paying for it, nothing :-) ) is based on some data points:
>
> > > 1. The last blog post on the Prototype blog was 10 months ago.
>
> > > 2. The library has had basically no activity since v1.7 was released
> > > November 2010.
>
> > > 3. The most recent three releases were:
> > > š šv1.7 - November 2010
> > > š šv1.6.1 - September 2009
> > > š šv1.6.0.3 - September 2008
>
> > > Three releases, in total, including "dot" releases, in the last three
> > > years. Compare with 16 releases (four major ones) of jQuery in that
> > > time period (v1.3 through v1.6.4).
>
> > > 4. Absolute use and trends:
>
> > > Absolute use:http://trends.builtwith.com/javascript
> > > Prototype trend:http://trends.builtwith.com/javascript/Prototype
> > > jQuery trend:http://trends.builtwith.com/javascript/JQuery
>
> > > 5. Questions tagged on StackOverflow:
>
> > > jQuery: š š114,842
> > > Prototype: š 2,152

ncubica

unread,
Sep 23, 2011, 9:49:46 PM9/23/11
to Prototype & script.aculo.us
Hi TJ, I would like to help in the implementation of the UI interface
for prototype, who do you think can lead this feature? I'm think I'm
best using prototype than working on the core, I could help in this
stage, and also I would love a renaissance of protoype.

I think in the same way of you... we have to do something in a way of
help Andrew...

Best
Nahum

On Sep 23, 11:19 am, "T.J. Crowder" <t...@crowdersoftware.com> wrote:
> Hi,
>
> This was _JUST_ gone into, in depth:http://groups.google.com/group/prototype-scriptaculous/browse_thread/...

T.J. Crowder

unread,
Sep 25, 2011, 5:59:57 AM9/25/11
to Prototype & script.aculo.us
Hi Nahum,

On Sep 24, 2:49 am, ncubica <ncub...@gmail.com> wrote:
> Hi TJ, I would like to help in the implementation of the UI interface
> for prototype, who do you think can lead this feature?

It would be Andrew or some volunteer he nominates. I can't volunteer
for it, I'm over-committed and would just end up wasting people's
time.

Specifically with regard to UI, the first question will be whether
you're going to help Thomas with Script.aculo.us[1] (e.g., the
"official" UI-related stuff for Prototype), and/or help Sébastien,
Samuel, Yurij, and Vincent with Prototype UI[2], or help Ryan with
LivePipe[3], or (radical approach) port jQuery UI[4] to Prototype,
etc. (I'm sure the latter is a big project, but you get the benefit of
the fact jQuery UI is fairly broad and well-established, and the
jQuery UI team is a good size and has some funding.[5])

Prototype also really needs a skilled, engaged webmaster. Before his
other commitments required his time elsewhere, Tobie was working on
moving prototypejs.org to Github Pages[6], which would make updates to
it basically a matter of pushing a commit (or sending a pull request
for one).

But probably more than anything else, Prototype needs an engaged
leader again. The project lagged for a long time after Sam
(Stephenson, not Samuel Lebeau) went on to other things, and then he
finally officially handed the baton to Andrew and Tobie, but Tobie got
a new job and life changes, etc., and I don't know what (if any) time
he has for this, and Andrew's commented on his other demands. Is there
someone with sufficient knowledge, skill, and time to join them and
actively lead the project? I'd say doing it properly would require a
minimum of eight hours/week, every week: Fixing bugs, coordinating
fixes from other contributors, dealing with the website, determining
and actioning Prototype 2, etc., etc. It's a big commitment.

FWIW,
--
T.J. Crowder
Independent Software Engineer
tj / crowder software / com
www / crowder software / com

[1] http://madrobby.github.com/scriptaculous/contribute/
[2] http://prototype-ui.com/
[3] http://livepipe.net/
[4] http://jqueryui.com
[5] http://jqueryui.com/about
[6] http://pages.github.com/

clockworkgeek

unread,
Sep 26, 2011, 9:03:48 AM9/26/11
to Prototype & script.aculo.us
I also enjoy using Prototype's methodology and would gladly contribute
to keep it alive if the core team pull their weight too. For example
some pull requests on github are over a year old! In fact, I volunteer
to help clear the backlog. As one of the top prototype users on
stackoverflow (http://stackoverflow.com/tags/prototypejs/topusers) I
am capable of the necessary level of commitment. Please allow us all
to assist.

As to the future of the library I agree that jQuery has "won" the
battle by attrition but is woefully underachieving in the language
enhancements. I can imagine a stripped down version of Prototype that
has none of the DOM manipulation so it can coexist with all other
libraries and avoid dulicated features, and instead provide more of
those essentials like Enumerable and Class.

P.J.

unread,
Sep 26, 2011, 9:04:13 PM9/26/11
to Prototype & script.aculo.us
This! I second this!

> I can imagine a stripped down version of Prototype that
> has none of the DOM manipulation so it can coexist with all other
> libraries and avoid dulicated features, and instead provide more of
> those essentials like Enumerable and Class.

I've always seen Prototype JS as the library that tried to address the
needs of JS as a language. Wasn't that the motto or at least one of
the "features" touted when people started noticing it? Then jQuery
came about and people began wondering why they could do all these
fancy things with the DOM and UI that Prototype seemed lacking in.
With Scriptaculous and Scripty2 we've made decent progress; but I
honestly think that wagon has come and gone. When it comes to UI a
large number of people first turn to jQuery. When I have to think of
logic (how to do something) I try to think Prototype first, but if
it's at all dealing with the UI jQuery takes over. The problem with
duplicated functionality not only exists when it comes to the UI but
in other areas as well. Stuff like String#Include, AJAX, and selectors
are only some examples.

To recap: If Prototype JS can be re-written so it can coexist with all
other libraries and avoided duplicated features, and began to address
the native problems with the JS language, it has a very good chance of
picking up the attention it once had.

On Sep 26, 9:03 am, clockworkgeek <nonproffessio...@clockworkgeek.com>
wrote:

Gappa

unread,
Sep 27, 2011, 2:46:18 AM9/27/11
to Prototype & script.aculo.us
Hi all!

> To recap: If Prototype JS can be re-written so it can coexist with all
> other libraries and avoided duplicated features, and began to address
> the native problems with the JS language, it has a very good chance of
> picking up the attention it once had.

That would "almost" be http://documentcloud.github.com/underscore/ and/
or http://sugarjs.com/.

The idea itself is good! Not that sugar/underscore are bad, but both
do not cover all of the stuff that Prototype does.
And when working with jQuery one really missess all the great
Prototype magic and more logical API :)

Marty Amberg

unread,
Sep 27, 2011, 12:00:42 AM9/27/11
to prototype-s...@googlegroups.com
I was thinking today that besides better documentation, It be nice
having a site where one can go and download actual rl examples. I
could imagine a table type web page where one can view commands and
code snippets. Maybe it is out there somewhere already but I love to
see a place where if I wanted a popup for a page I could find one or
if I want to see a page full of classes processed and by clicking on
the command you could get all of this. There is no doubt that I am not
a coding genus and have problems sometimes grasping exactly how to do
something and end up googling till I find a good enough example. I
read someones comments from a while ago and the complaint was that the
documentation was there but very sparse. I can certainly donate the
server and space and time to the cause.
One more thing. Chevy and Ford and Honda, make lots of cars but that
does not mean they produce the best vehicles. It be much for fun
driving a Ferrari and they don't make as many. The point being, just
cause something is popular does not mean its the best and there are a
lot of other frameworks out there..

Thanks folks

T.J. Crowder

unread,
Sep 27, 2011, 5:07:40 AM9/27/11
to Prototype & script.aculo.us
On Sep 27, 5:00 am, Marty Amberg <mar...@edenstreet.com> wrote:
> I was thinking today that besides better documentation, It be nice
> having a site where one can go and download actual rl examples.

Yeah, I'd probably like to see either hosted examples on the
documentation pages themselves, or callouts to jsbin.com or
jsfiddle.net.

> One more thing.  Chevy and Ford and Honda, make lots of cars but that
> does not mean they produce the best vehicles.

Absolutely. But when a business is making a decision about a fleet
purchase, you can bet they're looking closely not only at the fuel
economy and reliability of the actual vehicles, but at the stability
and forecast longevity of the company behind them.

Walter Lee Davis

unread,
Sep 27, 2011, 9:58:44 AM9/27/11
to prototype-s...@googlegroups.com

On Sep 27, 2011, at 12:00 AM, Marty Amberg wrote:

> One more thing. Chevy and Ford and Honda, make lots of cars but that does not mean they produce the best vehicles. It be much for fun driving a Ferrari and they don't make as many. The point being, just cause something is popular does not mean its the best and there are a lot of other frameworks out there..

One talking point that got trotted out lots of times during the Dark Ages of Macintosh vs. Windows was that cockroaches were the largest population in numbers, but you wouldn't want one to write you a sonnet.

Walter

ncubica

unread,
Sep 29, 2011, 9:43:49 PM9/29/11
to Prototype & script.aculo.us
ok ... So T.J. I know you not have the ultimate decision for designate
a candidate for any position at Prototype community but certanly you
are the most near with the core team. So I would like to know if you
could email one of them and try to start designate person for
revitalize this community... I have a good widgets base with prototype
and I could build more... that's not a problem... but we need to
designate a OFFICIAL Site for hosted documentation and more important
EXAMPLES. scripteka is a good place to fine this, BUT doesn't have any
space for comments or is NOT organice in a way where you can easy find
the widget you are expect or more important ask to the community where
to find it or maybe build it by more experience developer,

I think is why jquery have grow that huge beside of the money behind
it. Cause a lot of designers with a lack knowledge of javascript can
use it... happend with php vs java... and a lot more examples out
there, We have to reinvent the way of how we communicate prototype...
to everybody out there!!!!

So please talk with them, I have a fully compromise for work with this
project a least 1 year... and this is because I explote my javascript
knowledge because this framework... and care about it.

Best
Ncubica...

Marty Amberg

unread,
Sep 29, 2011, 10:58:48 PM9/29/11
to prototype-s...@googlegroups.com
Might have been asleep during my previous google searches but I found
tj's site http://proto-scripty.wikidot.com today which seems like
answers questions of what I would want and think prototype needs to
grow. No sense in reinventing something that is there which is what
I ahd posted before. A place for solid examples and other useful things.

T.J. Crowder

unread,
Sep 30, 2011, 9:37:30 AM9/30/11
to Prototype & script.aculo.us
On Sep 30, 3:58 am, Marty Amberg <mar...@edenstreet.com> wrote:
> Might have been asleep during my previous google searches  but I found
> tj's  sitehttp://proto-scripty.wikidot.comtoday which seems like
> answers questions of what I would want  and think  prototype needs to
> grow.    No sense in reinventing something that is there which is what  
> I ahd  posted before.  A place for solid examples and other useful things.

Yeah, http://proto-scripty.wikidot.com is something people can use.
Not really "my" site, it's meant to be the community's. Before he
disappeared, Tobie was talking about migrating that content into the
new prototypejs.org site hosted on GitHub (where it's easy for anyone
to contribute via pull-request), which I think was and is a great
idea. WikiDot has gotten very aggressive with ads and most programmers
know enough markdown that it's familiar to them.

On Sep 30, 2:43 am, ncubica <ncub...@gmail.com> wrote:
> ok ... So T.J. I know you not have the ultimate decision for designate
> a candidate for any position at Prototype community but certanly you
> are the most near with the core team. So I would like to know if you
> could email one of them and try to start designate person for
> revitalize this community...

Andrew reads the list and of course anyone can send him an email
directly, but sure, I'll drop him a note calling his attention to this
thread.
Reply all
Reply to author
Forward
0 new messages