how to select an update channel

[Günter Zöchbauer]

I have read several times about update channels (stable, dev, bleeding edge).

I'm not able to build from source on my Debian x64 linux but want use the newest version available.

The URL I used for about 2 months seems not to get updated anymore (http://gsdview.appspot.com/dart-editor-archive-trunk/latest/darteditor-linux-64.zip)

How can I get more frequent updates than stable?

[Rico Wind]

https://www.dartlang.org/tools/editor/

There is a link to dev channel below the big stable channel download button, you should not rely on google cloud storage bucket layout

Cheers,

Rico

--
For other discussions, see https://groups.google.com/a/dartlang.org/
 
For HOWTO questions, visit http://stackoverflow.com/tags/dart
 
To file a bug report or feature request, go to http://www.dartbug.com/new

To unsubscribe from this group and stop receiving emails from it, send an email to misc+uns...@dartlang.org.

[Günter Zöchbauer]

Does Darteditor from dev channel know that it should update when a new dev channel release is available?

The link I posted above had also a version file in the same directory so I could build a script that checked if a new version is available and update automatically. 

[Rico Wind]

If you download a dev channel build of the editor it will know how to update when there is a new release on dev channel, if you download a stable channel build of the editor it will now how to update when there is a new release on stable channel :-)

Currently these are on the same version number (modulo the revision number baked in to some version reporting), i.e., currently both dev and stable are on 1.0.0.3. Moving forward dev channel will be something like the integration builds we offered before, i.e., weekly or biweekly releases - stable channel will only get critical bug fixes and otherwise be updated to a new minor version with something like 6 week intervals (this is not a guarantee, some releases may come faster some may come slower). Dev channel will, as the name indicates, have a higher probability of containing bugs, but you will get access to new features/performance improvements faster.

I hope that makes sense.

Cheers,

Rico

[dangli...@gmail.com]

The next version will be only the derivation of previous version or new incompatible version (2.0) will be introduced as in development phase?

[Rico Wind]

The next obvious version would be 1.1, there is already ton of stuff committed on bleeding edge that did not make it to 1.0. 2.0 is way out there in the future, we don't have anything to share about when that will be available - right now the focus is on increasing performance and stability of the current system, plus making sure developers have the tools and tool-features they need.

Cheers,

Rico

On Thu, Nov 14, 2013 at 11:55 AM, <dangli...@gmail.com> wrote:

The next version will be only the derivation of previous version or new incompatible version (2.0) will be introduced as in development phase?

--

[dangli...@gmail.com]

>> right now the focus is on increasing performance and stability of the current system.

You think that increasing performance is more important than introducing new language features?
That means that you want to say that the Dart like the programming language is already quite well off and there is no need to focus on its further improvement?
I certainly understand that the increase in speed is the most important task, but at the same time, can exists different ways to achieve the desired effect.
In particular, due to the introduction of new concepts and approaches in the programming language itself.
It is quite natural that the performance does not improve due to the limitations of language itself.
That is, even the programmer in most cases must not worry about how predictable will be translation source code in into RTTI. But in some cases programmer must have a possibility to make reasonable choice based on obvious assumptions that the selected way based on the language features and limitations may be considered as the best.

For example, a method of increasing the speed of calculation of integer operations.
Strict type specification in most cases will be faster than predictions of RTTI based on some conclusions during computations.

There are other comments but I will not be touching them now, so just answering your comment.

[Rico Wind]

On Thu, Nov 14, 2013 at 1:59 PM, <dangli...@gmail.com> wrote:

>> right now the focus is on increasing performance and stability of the current system.

You think that increasing performance is more important than introducing new language features?

I think it is important to let the dust settle on 1.0 and actually get some real large scale usage so that anything that would be added to 2.0 would be well thought through. So at the current time, yes, I think improving stability and performance is more important then introducing new language features.

 

That means that you want to say that the Dart like the programming language is already quite well off and there is no need to focus on its further improvement?

Yes, I think is already quite well off. I don't think I said anything about us not considering improvements, but you have to see this in the bigger perspective, you have to draw a line and not break people.

 

I certainly understand that the increase in speed is the most important task, but at the same time, can exists different ways to achieve the desired effect.
In particular, due to the introduction of new concepts and approaches in the programming language itself.
It is quite natural that the performance does not improve due to the limitations of language itself.

We still have ton of stuff that we can do to improve the performance on the existing system

 

That is, even the programmer in most cases must not worry about how predictable will be translation source code in into RTTI. But in some cases programmer must have a possibility to make reasonable choice based on obvious assumptions that the selected way based on the language features and limitations may be considered as the best.

For example, a method of increasing the speed of calculation of integer operations.
Strict type specification in most cases will be faster than predictions of RTTI based on some conclusions during computations.

There are other comments but I will not be touching them now, so just answering your comment.

--

[dangli...@gmail.com]

@Rico Wind
Thanks.
After the your exhaustively and pointedly answer  as "ton of stuff" more not required the additional information to understand exactly what you had in mind. For many, but not for me.

[Günter Zöchbauer]

This is exactly what I wanted to know.

Thank you!

[dangli...@gmail.com]

@Rico Wind

>> So at the current time, yes, I think improving stability and performance is more important then introducing new language features.

You says: "So at the current time, yes, I think improving stability more important".
Your words sounds like:
"So at the current time Dart is not stable and improving stability more important".
But why and how you want improving stability in stable version?
Very strange words. This means at least one of the following.
1. Dart 1.0 is unstable
2. Dart 1.0 is not Dart 1.0

[Rico Wind]

If you would stop paraphrasing my sentences that would be great (if you could also try to not extrapolate what I mean on stuff I have not been stating that would be even better).

There are bugs open in the bug-tracker, feel free to go glance at those, a lot of that has to be fixed. In a system with the size and complexity of dart there will also be new stuff coming in after a 1.0 launch - that needs to be fixed as well.

--

[dangli...@gmail.com]

>> If you would stop paraphrasing my sentences that would be great (if you could also try to not extrapolate what I mean on stuff I have not been stating that would be even better).

Paraphrasing your sentences? I not want this.
I'm a human and I do the logical conclusions. If you're talking about fire, I can say: "the object about you said is a warm object". Although you did not say it explicitly.
If it is fire this means that it is a warn even you don't want to say that fire is warm.
If for something required improving stability (as more important then for other) then my (only my?) logical conclusions in that the object (about which you say) is not stable.

>> a system with the size and complexity of dart there will also be new stuff coming in after a 1.0 launch - that needs to be fixed as well.

Current Dart version is 1.0.1
Current milestone in issue tracker is 1.1.

The milestone M9 is before milestone 1.1.

1. How Dart can release version 1.0.1 with opened 243 issues for milestone M9?

2. That is, stable Dart 1.0 with nice addition in 243 opened issues for milestone M9?

3. And immediately there is a new milestone 1.1?

Then I have still one question.

4 . What is a milestone M9 and why their open issues in Dart 1.0?

[dangli...@gmail.com]

OK. All fine. Just a bit of skepticism.
I hope that improving stability not means introducing breaking changes.

[jm]

Em sexta-feira, 15 de novembro de 2013 10h55min14s UTC-2, dangli...@gmail.com escreveu:

OK. All fine. Just a bit of skepticism.
I hope that improving stability not means introducing breaking changes.

Hi there dangli-X.

I'm just a random guy here, occasionally stumbling by. It strikes me that your skepticism here seems wildly unfounded and that you are on the edge of doing so quite often. Unless have the history of (are) Mezoni, whom I do feel got crushed rather unfairly (with no particular blame on anyone in particular, because of the way of the world, otherwise known as evil, took its toll on him).

But please, Please, do not suppose you're dealing with government officials here, who have nothing better to do then to poke fun at you and at your impotence in dealing with them. That certainly isn't the way to engage anyone here.

Except for a historically funny bias to make Dart resemble Java, everyone here is very much focused on improving the status quo, especially off this list, often taking precious time out to answer questions, yours, to then return to their job saddened by your dissatisfaction.

Here you take the time of some of the best in the business, so if you can make your comments interesting that'll delight them. And that'll help Dart. Not that it is necessary to delight, being to the point is generally enough. Though in a foreign language that may be hard to achieve.

Now if you are Mezoni I can understand your bitterness. I understood Mezoni as having the silent wish to work at this project, for Google, and tried to be establish a reputation for himself as an opinion maker, since that is all he could do, and was heavily interested in the project. In striving to be important, but being left without a way to establish a track-record, he tragically ended up competing, and went on being bitter about not getting recognition. Not being a native speaker aggravated the experience for everyone.

I'd wished Google had offered him an internship (yeah!) for him to prove his worth, taking him off this list in the process, and also taking him off the path of doom. That would have been a risk worth taking as Mezoni apparently was talented, but in a bad place.

FWI.

Dart 1.0... I had my doubts, but I do think Dart provides a cohesive experience, and as such is the anti-thesis of a language designed by a committee. A big thumbs up. Now all that is left, is to be the better Java you folks set out to create (...) is to up the performance ante. Then, if Go's history is any indication, Dart will take the world by storm, starting with the server. In 2 years time it'll suddenly dawn on people that PHP is dead, and that Dart is the cornerstone of a world of programming with much improved tooling, a multi-language paradigm and much, much, less migraine. The browser will finally be the better operating system. Congratulations on making history.

[dangli...@gmail.com]

>>  I understood Mezoni as having the silent wish to work at this project, for Google

I am not programmer. I am businessman. I have small shop. For me this just a entertainment.
But if Google or you pay me money I do not mind.


воскресенье, 17 ноября 2013 г., 22:58:44 UTC+6 пользователь jm написал: