Chromeless “inactive”?

[Phil Schaf]

hi list,

today i read on the mozilla labs page that chromeless is “inactive”.

could someone (e.g. +Pascal Finette) elaborate why this is so, and if this state is just temporary or if there is yet another successor in the pipes?

thanks, phil

[Mike Hanson]

Hi, Phil - 

The ideas and code for Chromeless have grown into the HTML5 Applications project.  You can read all about it at https://wiki.mozilla.org/Apps

If you grab the latest Nightly build of Firefox [1], it has support for installing an HTML5 Application as a native launchable application on Windows and Mac, with support for more APIs coming soon [2].

-Mike

[1] http://nightly.mozilla.org/

[2] https://wiki.mozilla.org/WebAPI

--
You received this message because you are subscribed to the Google Groups "mozilla-labs" group.
To view this discussion on the web visit https://groups.google.com/d/msg/mozilla-labs/-/uehkN9Z9H8EJ.
To post to this group, send email to mozill...@googlegroups.com.
To unsubscribe from this group, send email to mozilla-labs...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/mozilla-labs?hl=en.

[Marcio Galli]

Some of the ideas are in applications ( agree in part with Mike ) but
some aspects, IMHO, are in B2G too.

The part I am to pitch is attribute called mozbrowser

Because B2G is like chromeless ( from the aspect that the top main app
(gaia) is HTML ) it needs to use something so its apps can work. Its
apps are sort of inner-browsers since apps = html. So it was started
in the iframe and this ~new || modified ~ iframe needs to have things
like: stop, reload, among other things.

Some of my writings with some ideas like this from the chromeless (
active ) days:

http://geckonnection.posterous.com/enhancing-an-iframe-with-more-browser-behavio
http://geckonnection.posterous.com/browser-functions-in-the-html-based-browser

And a bit of ideas about sites that started using a bit of *browsing
fetching from their domains* via networking data exchange.

http://geckonnection.posterous.com/outer-browsing-escaping-the-domain-from-a-web

---

I think this last part is where things gets hot because sites ( think
apps ) wants to do extended browsing things. And this is hte space
that needs thinking ( due the secure membranes we need ) but Its also
a space where growth and creativity emerges.

So, in this angle, a use case is an app ( in HTML ) that has full
browser functions . I understand that *full* is complicated matter
because some dev may want file system write and so on. So what I think
it is really important, to be, let's say, lower priority in projects
like "mozbrowser" attribute ( but yet a priority ) is to think about
these cases.

I am assuming mozbrowser attribute may bounce back to desktop and
eventually this may fit in specification

m

[Jock Murphy]

Can I just say that I am getting a bit tired of this from Mozilla.  Mozilla starts some cool project and then the guy behind it moves on and all of a sudden the project is dead.  Or we have something like Prisim -> Chromeless -> ?? situation we have here.

If I go to the prism page I find out that it is inactive, but not that it was essentially replaced with Chromeless.  If I go to the Chromeless page I get the same kind of thing.  Just a notice that it is inactive and no mention of any kind of replacement.

But also you killed the project before there was a viable replacement.  

All I want is a nice easy way to either provide site specific browsers to my clients, or for a nice simple way for them to make them themselves.  And the latter is really important.  It doesn't matter if there are HTML5 APIs that allow one to package their own app this way, users still have a need to create site specific browsers themselves.

I know there are third party tools, but few are cross platform, and few are suitable for average users.

But considering Mozilla's habit of walking away from projects like this, it makes me very nervous about my use of Persona for my current app, I have a mitigation plan but it would make me happier if Mozilla would just stick with tools like Prism and evolve them as techologies move forward.

[Pascal Finette | Mozilla]

Hi Mock,

Just to be clear - Chromeless is/was a Mozilla Labs Experiment. It was always clearly marked as such and no promises were made about the long-term viability of the project. The very fact that Mozilla as an organization experiments in the open and not behind closed doors (as is the norm) is something to be celebrated IMHO.

Persona on the contrary is not an experiment. It started out as one - very similar to Chromeless and moved on (other than Chromeless) to be a full-fledged product build and supported by Mozilla. So rest assured - Persona will be with us and continue to evolve for the long term.

Warmly,
P

To view this discussion on the web visit https://groups.google.com/d/msg/mozilla-labs/-/UMc3VYUOQH0J.

[Jock Murphy]

First off my name is Jock, not Mock, though my wife occasionally disagrees  ;)

My point was that Prism wasn't experimental — at least not that I can see now from the wiki — and when the one primary guy left it died.  There wasn't a link to an alternative, there wasn't an explanation, it was just over.  Chromeless came along and while not exactly a replacement, and not an end user technology, I could get it to work for my customers.

Then the experiment ended, was that announced on the Mozilla Labs blog? Nope.  It just ended, and now there is a kind of replacement that is still in development, and there isn't a link to it on the Chromeless site.

I love that Mozilla experiments, and that they do it in public, but not everything that has died because someone left were experiments.  When this happens, it can make us nervous.  That is all I am saying.

The persona comment was really an aside, but despite assurances about Persona I am still nervous.  For example:

Recently I had to report a bug about the Persona website about a font rendering issue — an issue that was so obvious it shouldn't have been pushed live IMHO.  I as dismayed to find out that there was "contact us" link on the main page.  Yes you can click "Need Help" and then make a big guess that "Suggestion Box" might be the way to contact someone for help.  When I contacted the webmaster, I got a reply from a non-mozilla email address, and had to argue if it was bug or not.

If it had been one of my customers who tried to do that, none of that would have made them feel warm and fuzzy about using Persona.

[Pascal Finette | Mozilla]

Hi Jock,

First of all - my apologies. I surely didn't want to butcher your name!

In terms of Prism - the point is mute now (as Prism is a thing of the past), yet just to make the argument: Prism was always clearly marked as a Mozilla Labs experiment. Judging from your response I am sure we could and should have made that clearer.

My apologies if that caused pain or trouble for you. Prism also didn't disappear because Matthew Gertner "left" (Matthew was one of a couple of people working on Prism and actually a contractor) - the team working on Prism made the decision that it was time to move on, as the specific technological direction we pursued with Prism didn't seem to be the right one long-term. Matthew then went on and took the Prism code (which is open source), forked it and for a while ran a new project under the umbrella of his own company. He later stopped working on this project - I don't have the context why that happened.

Chromeless is interesting as we originally conceived it as a way to rapidly prototype new browser UI: Take an empty rendering shell, build some hooks and you can create an experimental browser UI using HTML/CSS/JS. The community thought differently (which is awesome and the power of community) and started hacking on Chromeless in a way which was more akin to Prism. We never intended the project to be this - and I believe never stated anything else than that Chromeless is a way to prototype browser-based UIs. The project ran it's course, we found it actually not to be too useful for the stated purpose and decided to focus our energy on other projects.

Now - this is the very long way of saying: I am personally very sorry if we caused confusion or even some pain for you and others. This was and is not our intention. We could and should have communicated better. It should not make you nervous about Mozilla at large though -- these things were always experiments, not projects. We take the projects very serious, commit to being a strong, stable and long-term partner for our community and the products they build on top of our work.

Warmly,

P

P.S.: For context - I used to run Mozilla Labs, the very part of the Mozilla project which put out Prism and Chromeless. Chromeless was conceived by me and Lloyd Hilaiel, one of our Staff Engineers. :)

To view this discussion on the web visit https://groups.google.com/d/msg/mozilla-labs/-/7FzBKEiLh4gJ.

[Jock Murphy]

Oh I was sure the name thing was a typo, you wouldn't be the first.  

Let me clarify something, when I say that the person working on something leaves and the project is dead, I am describing what it looks like on the outside.  There is quite the concordance between inactive projects and people no longer at Mozilla.  Aza Raskin leaves and seemingly everything he was doing stops, as just one example.  Even if the assumption is wrong, it becomes point that people jumpt to.  

As for Prism, take a look at its Wiki page there isn't a single mention of it being experimental.  And if you search for Prism that is the top hit for me.  Also notice that the top of that page says:

Mozilla Labs itself started a new project called Chromeless. See their blog post athttp://mozillalabs.com/blog/2011/02/prism-is-now-chromeless/

Its a little hard not to take that as saying that Chromeless was to be a replacement for Prism.  I take you at your word it wasn't intended to be, but that isn't the messaging I got.  

The other thing to remember is that in this post Web 2.0, perpetual beta, buzzword compatible world we live in; real serious and active products/projects are called Beta, called Experiments, and even called Hobbies.  Sadly (in this case) these words have evolved, so we now have experiment in the traditional sense, and experiment in the sense of something that is a kind of sporadically active side project, a useful tool for power users.  

The problem is that there is still a need for something like Prism — an end user tool for making site specific browsers — and that need is different from what Chromeless is/was, or what HTML5 applications aims to achieve.  All we are left with are platform specific tools that don't quite have the same feature set.  I can't tell you how big that need is, but I can tell you it exists.

There is exactly no need to dwell on this.  What is done is done.  I may well end up building some kind of tool do just what Prism did, I can't promise it will be Gecko based, but the niche will be filled.  I don't quite know where I will find the time, but my customers have the need, and I would prefer a generic solution over something custom just for my app.

It's just a shame because there used to be this great thing I could point people too... ;)

Thank you for the apology, it wasn't needed, but I deeply appreciate it.

Jock

[Jock Murphy]

Pascal— 

I am not sharing this with you as a bug report, I will find the proper venue to tod that, but just as a further indication of what makes me nervous about Persona.  I think there is an object lesson in there somewhere ;)

I noticed that if you click on the "Need Help?" link on Persona.org you are taken to a page that talks about BrowserID.  This would be a little confusing to novice users so I thought I would send feedback — and what do you know? There is a tab called "suggestion box" that is apparently for just that purpose.

But do you know what happens when I click on that and then click "Submit Your Feedback?"  I am taken to a page that tells me I don't have the latest version of Firefox, and that I need to download it before I can submit feedback.  I do have the latest version of Firefox, but that isn't what caused the problem: I was using Chrome at the time.

(BTW if you click on Products and Services, Persona isn't on that list.)

Now imagine it is some frustrated novice user who just wants to report a bug, or worse yet, has a problem logging in.  That is going to be a very negative experience for them.  I believe in what Mozilla is trying to do with Persona, that is why I am using it.  

As I say I will search and find the right venue for reporting this bug, sadly it isn't obvious.  But I just wanted to point out why — in the light of my other comments — I worry about my decision to use Persona.

If there is a lesson here, it is that the details matter.  Anyone on a project like Persona needs to be looking at all the way their site connects to other things and make sure there is a consistant message, and that there is a clear simple way to report issues.

[Benjamin Young]

Hey Mike,

Is there any information on how to use the "installable" apps feature in either Nightly or (if available) Firefox 15? I've tried to look this up several times, but have never found clear information on how to do it or even where to look to see if the feature is available in the FF I have installed.

It would be great if someone could update ye olde Prism page to point it to Chromeless and update the Chromeless page to point it to some clear explanation of either a) how to do the same sort of Single Site Browser "installation" and/or b) a roadmap of what's planned.

To get and keep the community at large engaged (in Labs as well as in actual "shipping" products) knowing what's next (or even hoped to be next) is key. That's pretty visible with Firefox, but piecing together random repos, wiki pages, "promo" pages, scattered blog posts, mailing list posts, etc gets wearisome and it quickly becomes easier to click "Create shortcut..." (or whatever) in Google Chrome...something I'm loath to do as I don't want to switch, but I do want site specific browsing so they show up in my task bar, alt-tab queue, and (ideally) can send me notifications.

I know (from research) that much of this is planned and even in progress, but I've still not been able to figure out how far along it is, when I (or anyone else) might be able to use it, or (and this is key for Mozilla) how I and others might contribute to making it happen, promoting it, or at the very least filing bugs.

Thanks again for linking to some related information.

Later,
Benjamin

On 4/24/12 12:27 PM, Mike Hanson wrote:

Hi, Phil -�

The ideas and code for Chromeless have grown into the HTML5 Applications project. �You can read all about it at�https://wiki.mozilla.org/Apps

If you grab the latest Nightly build of Firefox [1], it has support for installing an HTML5 Application as a native launchable application on Windows and Mac, with support for more APIs coming soon [2].

-Mike

[1]�http://nightly.mozilla.org/

[2]�https://wiki.mozilla.org/WebAPI

On Apr 24, 2012, at 9:23 AM, Phil Schaf wrote:

hi list,

today i read on the mozilla labs page that chromeless is �inactive�.

could someone (e.g. +Pascal Finette) elaborate why this is so, and if this state is just temporary or if there is yet another successor in the pipes?

thanks, phil

--
You received this message because you are subscribed to the Google Groups "mozilla-labs" group.
To view this discussion on the web visit https://groups.google.com/d/msg/mozilla-labs/-/uehkN9Z9H8EJ.
To post to this group, send email to mozill...@googlegroups.com.
To unsubscribe from this group, send email to mozilla-labs...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/mozilla-labs?hl=en.

[Bill Walker]

Benjamin,

http://www.mozilla.org/en-US/apps/partners/

The project Mike Hanson referred to back in April has grown into an initiative that will, among other things, deliver a marketplace of applications for Firefox OS. While we concentrate on supporting Firefox OS, we are also working to create a native App experience on Android, Mac, Windows, and Linux. The link about should lead you to lots of useful info about building HTML Apps and installing them.

Application support is currently enabled on Firefox 16, and is expected to remain enabled when Firefox 16 ships in October.

hope this helps,

-Bill

On Sep 26, 2012, at 5:54 AM, Benjamin Young wrote:

Hey Mike,

Is there any information on how to use the "installable" apps feature in either Nightly or (if available) Firefox 15? I've tried to look this up several times, but have never found clear information on how to do it or even where to look to see if the feature is available in the FF I have installed.

It would be great if someone could update ye olde Prism page to point it to Chromeless and update the Chromeless page to point it to some clear explanation of either a) how to do the same sort of Single Site Browser "installation" and/or b) a roadmap of what's planned.

To get and keep the community at large engaged (in Labs as well as in actual "shipping" products) knowing what's next (or even hoped to be next) is key. That's pretty visible with Firefox, but piecing together random repos, wiki pages, "promo" pages, scattered blog posts, mailing list posts, etc gets wearisome and it quickly becomes easier to click "Create shortcut..." (or whatever) in Google Chrome...something I'm loath to do as I don't want to switch, but I do want site specific browsing so they show up in my task bar, alt-tab queue, and (ideally) can send me notifications.

I know (from research) that much of this is planned and even in progress, but I've still not been able to figure out how far along it is, when I (or anyone else) might be able to use it, or (and this is key for Mozilla) how I and others might contribute to making it happen, promoting it, or at the very least filing bugs.

Thanks again for linking to some related information.

Later,
Benjamin

On 4/24/12 12:27 PM, Mike Hanson wrote:

Hi, Phil - 

The ideas and code for Chromeless have grown into the HTML5 Applications project.  You can read all about it at https://wiki.mozilla.org/Apps

If you grab the latest Nightly build of Firefox [1], it has support for installing an HTML5 Application as a native launchable application on Windows and Mac, with support for more APIs coming soon [2].

-Mike