Hello people, Here are a few critical issues due to bugs or lack of information or feature for porting Dr. Geo to P6. There were others critical issues from P6 but were resolved and will be hopefully integrated, when ? * Minimal Dr. Geo image <https://bugs.launchpad.net/bugs/1705661> * Integration with Playground <https://bugs.launchpad.net/bugs/1701460> * Export bitmap to clipboard <https://bugs.launchpad.net/bugs/1697418><-- it will be really cool to have. <https://bugs.launchpad.net/bugs/1697418> * Universal package for 64 bits <https://bugs.launchpad.net/bugs/1705668> Other issues: https://launchpad.net/drgeo/+milestone/17.09 My initial plane and need is to finish the porting and test by the end of August. Not sure it sounds realistic. Hilaire -- Dr. Geo http://drgeo.eu
Hilaire I do not understand what the issues that you need in Pharo 60. Stef On Fri, Jul 21, 2017 at 11:19 AM, Hilaire <hilaire@drgeo.eu> wrote:
Hello people,
Here are a few critical issues due to bugs or lack of information or feature for porting Dr. Geo to P6. There were others critical issues from P6 but were resolved and will be hopefully integrated, when ?
Minimal Dr. Geo image Integration with Playground Export bitmap to clipboard <-- it will be really cool to have. Universal package for 64 bits
Other issues: https://launchpad.net/drgeo/+milestone/17.09
My initial plane and need is to finish the porting and test by the end of August. Not sure it sounds realistic.
Hilaire
-- Dr. Geo http://drgeo.eu
Hi Stefane, Did you read the bug tickets in my previous email? Hilaire Le 21/07/2017 à 18:08, Stephane Ducasse a écrit :
I do not understand what the issues that you need in Pharo 60.
-- Dr. Geo http://drgeo.eu
Yes but there are not concrete enough for me :) On Fri, Jul 21, 2017 at 7:37 PM, Hilaire <hilaire@drgeo.eu> wrote:
Hi Stefane,
Did you read the bug tickets in my previous email?
Hilaire
Le 21/07/2017 à 18:08, Stephane Ducasse a écrit :
I do not understand what the issues that you need in Pharo 60.
-- Dr. Geo http://drgeo.eu
I really find it hard to believe !! Le 22/07/2017 à 12:04, Stephane Ducasse a écrit :
Yes but there are not concrete enough for me:)
-- Dr. Geo http://drgeo.eu
Hi, Can you detail your needs regarding the integration with Playground? Cheers, Doru
On Jul 21, 2017, at 11:19 AM, Hilaire <hilaire@drgeo.eu> wrote:
Hello people,
Here are a few critical issues due to bugs or lack of information or feature for porting Dr. Geo to P6. There were others critical issues from P6 but were resolved and will be hopefully integrated, when ?
⢠Minimal Dr. Geo image ⢠Integration with Playground ⢠Export bitmap to clipboard <-- it will be really cool to have. ⢠Universal package for 64 bits Other issues: https://launchpad.net/drgeo/+milestone/17.09
My initial plane and need is to finish the porting and test by the end of August. Not sure it sounds realistic.
Hilaire
-- Dr. Geo
-- www.tudorgirba.com www.feenk.com "It's not how it is, it is how we see it."
The problem is explained in the ticket: there is a bug in the position where the drgeo menu shows up when encapsulated in the Playground. It is not related to Glamour however. I may just decide to remove the Dr. Geo menu to avoid the bug. Le 21/07/2017 à 20:50, Tudor Girba a écrit :
Can you detail your needs regarding the integration with Playground?
-- Dr. Geo http://drgeo.eu
Aha. Thanks for the clarification. Doru
On Jul 21, 2017, at 8:59 PM, Hilaire <hilaire@drgeo.eu> wrote:
The problem is explained in the ticket: there is a bug in the position where the drgeo menu shows up when encapsulated in the Playground. It is not related to Glamour however. I may just decide to remove the Dr. Geo menu to avoid the bug.
Le 21/07/2017 à 20:50, Tudor Girba a écrit :
Can you detail your needs regarding the integration with Playground?
-- Dr. Geo http://drgeo.eu
-- www.tudorgirba.com www.feenk.com âLive like you mean it."
You are welcome. I also wrote here a few days ago about it. The bug is Morphic specific. Hilaire Le 22/07/2017 à 09:03, Tudor Girba a écrit :
Aha. Thanks for the clarification.
Doru
On Jul 21, 2017, at 8:59 PM, Hilaire<hilaire@drgeo.eu> wrote:
The problem is explained in the ticket: there is a bug in the position where the drgeo menu shows up when encapsulated in the Playground. It is not related to Glamour however. I may just decide to remove the Dr. Geo menu to avoid the bug.
-- Dr. Geo http://drgeo.eu
On 21. Jul 2017, at 11:19, Hilaire <hilaire@drgeo.eu> wrote:
Hello people,
Hi!
Here are a few critical issues due to bugs or lack of information or feature for porting Dr. Geo to P6. There were others critical issues from P6 but were resolved and will be hopefully integrated, when ?
⢠Minimal Dr. Geo image
I am concerned with the growth from Pharo3->Pharo6 as well and triggered by your mail looked at it again. In contrast to ImageCleaner>>#cleanUpForRelease I want to keep the Monticello/Metacello packages for now (and don't want to unload them after test). I call this with "pharo Image eval clean.st" and for whatever reason if I pass --save bad things (can't talk to mongod) happen during my system test but that is another story (maybe the save is executed at the _next_ image start again). World closeAllWindowsDiscardingChanges. (RPackage organizer packages select: [:package | package packageName includesSubstring: 'Test']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'Versionner']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'ProfStef']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'Ice']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'BaselineOf']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'ConfigurationOf']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName endsWith: '-Help']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName endsWith: 'Examples']) do: [:each | each removeFromSystem ]. ImageCleaner new cleanUpForRelease. Smalltalk snapshot: true andQuit: true.
Hi holger "the growth from Pharo3->Pharo6" us too :) this is why we invested in the bootstrap and this is why we will remove (and we started) packages and classes. And this is also why we will continue to repackage the system. Stef On Sat, Jul 22, 2017 at 10:43 AM, Holger Freyther <holger@freyther.de> wrote:
On 21. Jul 2017, at 11:19, Hilaire <hilaire@drgeo.eu> wrote:
Hello people,
Hi!
Here are a few critical issues due to bugs or lack of information or feature for porting Dr. Geo to P6. There were others critical issues from P6 but were resolved and will be hopefully integrated, when ?
⢠Minimal Dr. Geo image
I am concerned with the growth from Pharo3->Pharo6 as well and triggered by your mail looked at it again. In contrast to ImageCleaner>>#cleanUpForRelease I want to keep the Monticello/Metacello packages for now (and don't want to unload them after test).
I call this with "pharo Image eval clean.st" and for whatever reason if I pass --save bad things (can't talk to mongod) happen during my system test but that is another story (maybe the save is executed at the _next_ image start again).
World closeAllWindowsDiscardingChanges. (RPackage organizer packages select: [:package | package packageName includesSubstring: 'Test']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'Versionner']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'ProfStef']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'Ice']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'BaselineOf']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName beginsWith: 'ConfigurationOf']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName endsWith: '-Help']) do: [:each | each removeFromSystem ]. (RPackage organizer packages select: [:package | package packageName endsWith: 'Examples']) do: [:each | each removeFromSystem ].
ImageCleaner new cleanUpForRelease. Smalltalk snapshot: true andQuit: true.
On 22. Jul 2017, at 12:07, Stephane Ducasse <stepharo.self@gmail.com> wrote:
Hi holger
Hey Stef!
"the growth from Pharo3->Pharo6" us too :) this is why we invested in the bootstrap and this is why we will remove (and we started) packages and classes. And this is also why we will continue to repackage the system.
yes this is very fascinating for Pharo7. It would be nice if we could track size (image+changes but also ram usage, object count) over time to see our progress. I recently added ordinary metric upload to my "bob-bench.org" test and metric tracking. ;) holger
Indeed. I dreamed about the bootstrap in 2002 :) so now I enjoy it :) What I can tell you is that we will do another pass because we want Pharo core around 200 k + 300 k for a slow vm. And we will have one guy working on it during his PhD. We will continue to push with Pavel, christophe and guille too. Stef On Wed, Jul 26, 2017 at 9:15 PM, Holger Freyther <holger@freyther.de> wrote:
On 22. Jul 2017, at 12:07, Stephane Ducasse <stepharo.self@gmail.com> wrote:
Hi holger
Hey Stef!
"the growth from Pharo3->Pharo6" us too :) this is why we invested in the bootstrap and this is why we will remove (and we started) packages and classes. And this is also why we will continue to repackage the system.
yes this is very fascinating for Pharo7. It would be nice if we could track size (image+changes but also ram usage, object count) over time to see our progress. I recently added ordinary metric upload to my "bob-bench.org" test and metric tracking. ;)
holger
I don't share your enthusiasm. I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me). Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both. Hilaire Le 27/07/2017 à 21:28, Stephane Ducasse a écrit :
Indeed. I dreamed about the bootstrap in 2002:) so now I enjoy it:) What I can tell you is that we will do another pass because we want Pharo core around 200 k + 300 k for a slow vm. And we will have one guy working on it during his PhD. We will continue to push with Pavel, christophe and guille too.
-- Dr. Geo http://drgeo.eu
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote:
I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
âI have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff. There is no simple solution. One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process. Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases. âThis is why we all need âin the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community. We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement. Regards, -- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
Just to chip in - I hope we can get past this bump (and I think it is a bump). I am actually ecstatic that recent improvements - a 64bit vm, a minimal image and git integration, have opened the door to Pharo working on AWS Lambda. Also having a CI system to help crank these things out is also a big win. So I applaud the work, progress and vision that has got us here ! All large systems accrue infrastructure and unexpected dependencies that take effort to modernise. I don't think this is solely a Pharo thing. (I just recently worked with a Java team that used SnapCI for builds, and when it was shut down it was 6+ weeks to migrate off it and stabilise things). We are close - we just need to rally around to get a good stable base that can support the next round of revolution (aka P7). Tim Sent from my iPhone
On 28 Jul 2017, at 11:00, Serge Stinckwich <serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote: I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
âI have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
âThis is why we all need âin the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
Changing too many things at once is indeed annoying. Now, I am ready to live with that but at one point, I think that we will have to move to something like I see done in other fast evolving ecosystems. In Hadoop for example, Hortonworks (a distribution) moved to a set of slow evolving substrate that is stable and know to stay stable for a long period (HDFS, YARN) and a set fast moving releases for projects that do build on top (Spark). Holding back on the new things makes you feel like you use a tool of the past. Living on the bleeding edge is not doable because you need to solve too many non business centric issues. There needs to be a combination. As far as I am concerned, I worked in 3.0 a lot, skipped the whole 4.0 ship, embarked on the 5.0 and, albeit if I did a bit on 6.0, I may not develop production code on it at the moment. 7.0 looks okay but there are lot of changing things there, so, that is also too much for me. 6.1 can lure me in with Iceberg and 64-bit UFFI and fast inspectors on large collections. I need a platform I can understand and build upon. There needs to be a semblance of LTS in this. Maybe a 6.1, 6.2, 6.3 story and a 7.x line with boostrap magic and what not. 6.x is a great platform and has a lot going for it if stable enough. I have projects coming my way and using Pharo is an option. Now, I need something that is not going to shift under my feet. Especially if I want to embark crew along. Phil On Fri, Jul 28, 2017 at 11:00 AM, Serge Stinckwich < serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote:
I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
âI have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
âThis is why we all need âin the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
On 28 Jul 2017, at 15:13, phil@highoctane.be wrote:
Changing too many things at once is indeed annoying.
Now, I am ready to live with that but at one point, I think that we will have to move to something like I see done in other fast evolving ecosystems.
In Hadoop for example, Hortonworks (a distribution) moved to a set of slow evolving substrate that is stable and know to stay stable for a long period (HDFS, YARN) and a set fast moving releases for projects that do build on top (Spark).
Holding back on the new things makes you feel like you use a tool of the past. Living on the bleeding edge is not doable because you need to solve too many non business centric issues.
There needs to be a combination.
As far as I am concerned, I worked in 3.0 a lot, skipped the whole 4.0 ship, embarked on the 5.0 and, albeit if I did a bit on 6.0, I may not develop production code on it at the moment. 7.0 looks okay but there are lot of changing things there, so, that is also too much for me.
6.1 can lure me in with Iceberg and 64-bit UFFI and fast inspectors on large collections. I need a platform I can understand and build upon.
There needs to be a semblance of LTS in this.
But even the LTS concept does not solve all problems. Every 2 years there is a new LTS, which is supported for 5 years. But in most projects (the same happened here), you are lazy and wait 5 years until you *have* to upgrade. And by then the difference between what you started with (say 12.04) and the current stable (say 16.04) can already be huge (remember, you skipped one LTS release) and the next one is already coming (18.04). Change is unavoidable, it is not just part of life, it *is* life.
Maybe a 6.1, 6.2, 6.3 story and a 7.x line with boostrap magic and what not.
6.x is a great platform and has a lot going for it if stable enough.
I have projects coming my way and using Pharo is an option. Now, I need something that is not going to shift under my feet.
Especially if I want to embark crew along.
Phil
On Fri, Jul 28, 2017 at 11:00 AM, Serge Stinckwich <serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote: I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
âI have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
âThis is why we all need âin the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
On Fri, Jul 28, 2017 at 3:25 PM, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 28 Jul 2017, at 15:13, phil@highoctane.be wrote:
Changing too many things at once is indeed annoying.
Now, I am ready to live with that but at one point, I think that we will have to move to something like I see done in other fast evolving ecosystems.
In Hadoop for example, Hortonworks (a distribution) moved to a set of slow evolving substrate that is stable and know to stay stable for a long period (HDFS, YARN) and a set fast moving releases for projects that do build on top (Spark).
Holding back on the new things makes you feel like you use a tool of the past. Living on the bleeding edge is not doable because you need to solve too many non business centric issues.
There needs to be a combination.
As far as I am concerned, I worked in 3.0 a lot, skipped the whole 4.0 ship, embarked on the 5.0 and, albeit if I did a bit on 6.0, I may not develop production code on it at the moment. 7.0 looks okay but there are lot of changing things there, so, that is also too much for me.
6.1 can lure me in with Iceberg and 64-bit UFFI and fast inspectors on large collections. I need a platform I can understand and build upon.
There needs to be a semblance of LTS in this.
But even the LTS concept does not solve all problems.
Every 2 years there is a new LTS, which is supported for 5 years.
You are a Ubuntu user. I am a CentOS|RHEL user. There support is 10 years. Check this: https://wiki.centos.org/About/Product Official party line here: https://access.redhat.com/support/policy/updates/errata TL;DR: CentOS6: until 2020 for maintenance updates CentOS7, until 2024 for maintenance updates. Using the recent/latest is possible with e.g. * Software Collections https://www.softwarecollections.org/en/ * LXC If one needs a 4.x kernel that is doable too via elrepo. But there is a lot of backport happening by RedHat. There is a reason these guys are making gobs of money: they support business. Frankly I wouldn't touch Ubuntu with a 10 feet pole anymore now that I have 100+ servers running CentOS under management.
But in most projects (the same happened here), you are lazy and wait 5 years until you *have* to upgrade.
And by then the difference between what you started with (say 12.04) and the current stable (say 16.04) can already be huge (remember, you skipped one LTS release) and the next one is already coming (18.04).
To be frank 14.04 is the only decent release I know. All of the other ones have issues of one kind or another.
Change is unavoidable, it is not just part of life, it *is* life.
Sure. But change is to be managed and the business is what makes money. So, not catering to business needs is a losing proposition. There is this business by Clement and Esteban that apparently will come to life at one point. I guess business forces may have more impact at that time frame. Wait and see. Phil
Maybe a 6.1, 6.2, 6.3 story and a 7.x line with boostrap magic and what not.
6.x is a great platform and has a lot going for it if stable enough.
I have projects coming my way and using Pharo is an option. Now, I need something that is not going to shift under my feet.
Especially if I want to embark crew along.
Phil
On Fri, Jul 28, 2017 at 11:00 AM, Serge Stinckwich < serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote: I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
âI have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
âThis is why we all need âin the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
+ 100 On Fri, Jul 28, 2017 at 3:25 PM, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 28 Jul 2017, at 15:13, phil@highoctane.be wrote:
Changing too many things at once is indeed annoying.
Now, I am ready to live with that but at one point, I think that we will have to move to something like I see done in other fast evolving ecosystems.
In Hadoop for example, Hortonworks (a distribution) moved to a set of slow evolving substrate that is stable and know to stay stable for a long period (HDFS, YARN) and a set fast moving releases for projects that do build on top (Spark).
Holding back on the new things makes you feel like you use a tool of the past. Living on the bleeding edge is not doable because you need to solve too many non business centric issues.
There needs to be a combination.
As far as I am concerned, I worked in 3.0 a lot, skipped the whole 4.0 ship, embarked on the 5.0 and, albeit if I did a bit on 6.0, I may not develop production code on it at the moment. 7.0 looks okay but there are lot of changing things there, so, that is also too much for me.
6.1 can lure me in with Iceberg and 64-bit UFFI and fast inspectors on large collections. I need a platform I can understand and build upon.
There needs to be a semblance of LTS in this.
But even the LTS concept does not solve all problems.
Every 2 years there is a new LTS, which is supported for 5 years.
But in most projects (the same happened here), you are lazy and wait 5 years until you *have* to upgrade.
And by then the difference between what you started with (say 12.04) and the current stable (say 16.04) can already be huge (remember, you skipped one LTS release) and the next one is already coming (18.04).
Change is unavoidable, it is not just part of life, it *is* life.
Maybe a 6.1, 6.2, 6.3 story and a 7.x line with boostrap magic and what not.
6.x is a great platform and has a lot going for it if stable enough.
I have projects coming my way and using Pharo is an option. Now, I need something that is not going to shift under my feet.
Especially if I want to embark crew along.
Phil
On Fri, Jul 28, 2017 at 11:00 AM, Serge Stinckwich <serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote: I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
I have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
This is why we all need in the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
Hilaire We were discussing this with Esteban and he will do a stab at making sure that we can deploy desktop applications much easier with Pharo. Because we need it. Esteban deployed several desktop apps in the past and we should leverage this knowledge. I also wants to write this in a forthcoming tutorial. Stef On Sun, Jul 30, 2017 at 4:20 PM, Stephane Ducasse <stepharo.self@gmail.com> wrote:
+ 100
On Fri, Jul 28, 2017 at 3:25 PM, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 28 Jul 2017, at 15:13, phil@highoctane.be wrote:
Changing too many things at once is indeed annoying.
Now, I am ready to live with that but at one point, I think that we will have to move to something like I see done in other fast evolving ecosystems.
In Hadoop for example, Hortonworks (a distribution) moved to a set of slow evolving substrate that is stable and know to stay stable for a long period (HDFS, YARN) and a set fast moving releases for projects that do build on top (Spark).
Holding back on the new things makes you feel like you use a tool of the past. Living on the bleeding edge is not doable because you need to solve too many non business centric issues.
There needs to be a combination.
As far as I am concerned, I worked in 3.0 a lot, skipped the whole 4.0 ship, embarked on the 5.0 and, albeit if I did a bit on 6.0, I may not develop production code on it at the moment. 7.0 looks okay but there are lot of changing things there, so, that is also too much for me.
6.1 can lure me in with Iceberg and 64-bit UFFI and fast inspectors on large collections. I need a platform I can understand and build upon.
There needs to be a semblance of LTS in this.
But even the LTS concept does not solve all problems.
Every 2 years there is a new LTS, which is supported for 5 years.
But in most projects (the same happened here), you are lazy and wait 5 years until you *have* to upgrade.
And by then the difference between what you started with (say 12.04) and the current stable (say 16.04) can already be huge (remember, you skipped one LTS release) and the next one is already coming (18.04).
Change is unavoidable, it is not just part of life, it *is* life.
Maybe a 6.1, 6.2, 6.3 story and a 7.x line with boostrap magic and what not.
6.x is a great platform and has a lot going for it if stable enough.
I have projects coming my way and using Pharo is an option. Now, I need something that is not going to shift under my feet.
Especially if I want to embark crew along.
Phil
On Fri, Jul 28, 2017 at 11:00 AM, Serge Stinckwich <serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote: I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
I have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
This is why we all need in the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
Changes is ok. What annoying is announced new feature of P6, but not palpable for common mortal as me (whose main occupation is teaching to teenagers and programming a hobby): * Pharo can now be bootstrapped from source code managed by Git Will a new Pharo user entering Pharo programming get any chance to understand or use it? I don't and it is frustrating. Hilaire Le 28/07/2017 à 15:13, phil@highoctane.be a écrit :
Changing too many things at once is indeed annoying.
-- Dr. Geo http://drgeo.eu
We understand that perfectly. Now for us we get money now and may be not the same in the future. Then there are the things that - must be done and consume a lot: 64bits, uFFI, - super important to do git - also super important: cleaning and improving the system. So we pay attention for backward compatibilities but we cannot cover everything. Stef On Fri, Jul 28, 2017 at 3:13 PM, phil@highoctane.be <phil@highoctane.be> wrote:
Changing too many things at once is indeed annoying.
Now, I am ready to live with that but at one point, I think that we will have to move to something like I see done in other fast evolving ecosystems.
In Hadoop for example, Hortonworks (a distribution) moved to a set of slow evolving substrate that is stable and know to stay stable for a long period (HDFS, YARN) and a set fast moving releases for projects that do build on top (Spark).
Holding back on the new things makes you feel like you use a tool of the past. Living on the bleeding edge is not doable because you need to solve too many non business centric issues.
There needs to be a combination.
As far as I am concerned, I worked in 3.0 a lot, skipped the whole 4.0 ship, embarked on the 5.0 and, albeit if I did a bit on 6.0, I may not develop production code on it at the moment. 7.0 looks okay but there are lot of changing things there, so, that is also too much for me.
6.1 can lure me in with Iceberg and 64-bit UFFI and fast inspectors on large collections. I need a platform I can understand and build upon.
There needs to be a semblance of LTS in this.
Maybe a 6.1, 6.2, 6.3 story and a 7.x line with boostrap magic and what not.
6.x is a great platform and has a lot going for it if stable enough.
I have projects coming my way and using Pharo is an option. Now, I need something that is not going to shift under my feet.
Especially if I want to embark crew along.
Phil
On Fri, Jul 28, 2017 at 11:00 AM, Serge Stinckwich <serge.stinckwich@gmail.com> wrote:
On Fri, Jul 28, 2017 at 9:34 AM, Hilaire <hilaire@drgeo.eu> wrote:
I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
I have sometimes the same concerns with Pharo or some tools of the Pharo ecosystem. I know that we are trying to do our best and regarding the number of core developers we have already an incredible platform. But sometimes, you need to very simple updates and because of subtle problems with VM/configurations/CI/ etc ... this is not that simple and we need to spend times on boring stuff.
There is no simple solution.
One solution might be that the core developers only focus on core Pharo functionalities but I think this is somewhat difficult, because most of the dev are from RMOD. RMOD is a research unit and could not spend all his money/effort on an engineering process.
Another solution is to grow our community. More people, more companies to sustain more engineers through the consortium. The more people we are able to attract, the more people will help to develop working solutions for problems like deployment or to have bug-fixing intermediate releases.
This is why we all need in the community to do as much as possible advertisements: lectures at universities, talk to your colleague about Pharo, do demos in companies, at open-source forums, use Twitter do talk about Pharo ecosystem, the software you are developing with Pharo. Don't hide problems but talk about our nice platform and our community.
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
Regards,
-- Serge Stinckwich UCN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/
Serge, The only reason I decided to port DrGeo from Squeak to Pharo is because the project wanted to be business centric with features evolution. Changes was not annoying me, it is one reason I decided to switch to Pharo. But still after many years, I find Pharo frustrating for one willing to develop and deploy a desktop application. Hilaire Le 28/07/2017 à 11:00, Serge Stinckwich a écrit :
We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
-- Dr. Geo http://drgeo.eu
The main problem is DrGeo is one a few software in the community to be deployed as a desktop app. All the others are dev software (Calypso, Roassal, MOOSE) or web-based app that don't need specific work for deployment. I need to deployed simulation engines that we are building with my team outside the community to people who are not CS experts, so I guess I will need something similar to DrGeo or use a web app. Envoyé de mon iPhone
Le 28 juil. 2017 à 18:43, Hilaire <hilaire@drgeo.eu> a écrit :
Serge,
The only reason I decided to port DrGeo from Squeak to Pharo is because the project wanted to be business centric with features evolution. Changes was not annoying me, it is one reason I decided to switch to Pharo. But still after many years, I find Pharo frustrating for one willing to develop and deploy a desktop application.
Hilaire
Le 28/07/2017 à 11:00, Serge Stinckwich a écrit : We have done this with Stephane in the early days of Pharo at open-source forums in France and I remember that you come in the community after we meet you in one of these forums :-) So DrGeo2 exists because of this kind of advertisement.
-- Dr. Geo http://drgeo.eu
Hilaire We invested 3 years of PhD of Guillermo + 8 months of hard work of christophe and guillermo. Pavel spent years to produce the mini image and now with the bootstrap it is opening a lot of possibilities. We will do it and it will work. And ***I LOVE THEIR WORK***. I DREAMED about it more than 15 years ago and they make it true. This is a game changer. It forces the system to be a lot more modular and cleaner. In the future people will take a core and their own configuration and build whatever they want. And we will build Pharo based on this core. I understand that you want to focus on DrGeo but the situation is like that. There is no magic. Stay in Pharo 30 but do not expect people will help you because Pharo is now more than 4 years old. I think that Pavel was really supportive with you. Now when I see your emails, I often see you ranting and people takes on their free time to help you so the more you rant the less people will help you. Ask yourself when it is the last time that you fixed a typo in Pharo and proposed a fix or just reproduce a bug that is not on your direct interest. You see people are sensible to this matter. I imagine that you know the soup of stone story. Pharo is our common goods and this is important that we all pay attention to it. Stef On Fri, Jul 28, 2017 at 10:34 AM, Hilaire <hilaire@drgeo.eu> wrote:
I don't share your enthusiasm.
I once set up a satisfactory build environment for DrGeo, based on P3. As long as I stay with P3, I can concentrate on DrGeo code: write the code, then fire up a build script to deploy the application. Now porting to P6 is a pain: the infrastructure to deploy a desktop application has not evolve since P3, I have to build again a deployment environment from scratch (VM support, shrinked/built image, I don't know the promise of minimal image build up is not palpable for me).
Now If I have to spend days on that, I am not sure I will do it again, I can't compete against other geometry application if I have to fight against pharo too. What I want is to concentrate working on DrGeo not Pharo, sorry to make it explicit but I can't much offer to do both.
Hilaire
Le 27/07/2017 à 21:28, Stephane Ducasse a écrit :
Indeed. I dreamed about the bootstrap in 2002 :) so now I enjoy it :) What I can tell you is that we will do another pass because we want Pharo core around 200 k + 300 k for a slow vm. And we will have one guy working on it during his PhD. We will continue to push with Pavel, christophe and guille too.
-- Dr. Geo http://drgeo.eu
Hi Holger, With Pharo3 I was proceeding like this, by uninstallation of packages. However there is the promise of building from a minimal image, but it is not documented AFAIK. By curiosity, what is the size of your resulting image. Hilaire Le 22/07/2017 à 10:43, Holger Freyther a écrit :
I am concerned with the growth from Pharo3->Pharo6 as well and triggered by your mail looked at it again. In contrast to ImageCleaner>>#cleanUpForRelease I want to keep the Monticello/Metacello packages for now (and don't want to unload them after test).
I call this with "pharo Image eval clean.st" and for whatever reason if I pass --save bad things (can't talk to mongod) happen during my system test but that is another story (maybe the save is executed at the_next_ image start again).
-- Dr. Geo http://drgeo.eu
On 22. Jul 2017, at 14:00, Hilaire <hilaire@drgeo.eu> wrote:
Hi Holger,
With Pharo3 I was proceeding like this, by uninstallation of packages. However there is the promise of building from a minimal image, but it is not documented AFAIK.
By curiosity, what is the size of your resulting image.
I see 31mb image size for Pharo3 and 49mb for Pharo6. The changes file is about the same. From the top command on a test system. "pharo" is a Pharo6 VM and image and pharo-vm is the pharo3 system. PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 4206 root 20 0 1056684 74036 1600 S 5.7 1.2 1582:15 pharo-vm 10459 root 20 0 104040 90348 2124 S 5.0 1.5 369:00.29 pharo I looked a bit at memory usage and see if some random changes make a difference * Shrink ZnMimeType ExtensionMaps to a minimum when not handling extensions? * ASTCache reset to get rid of the spotter cache. Not sure if the SessionManager runs ASTCache>>#shutDown properly. It looks like it should but it doesn't? * ByteString compaction? I did strs := ByteString allInstances. str -> strs asSet. The numbers on a not fully clean image are 99874 vs. 42139. I wonder in which hell I end if I use >>#becomeForward: to compact it (I tried this once for GNU Smalltalk and I am sure others tried it before). I think anyone using == on String is using the wrong class? Any good arguments against doing it? I am moving my Pharo6 image into production right now and then will look into setting up CI for Pharo7 with a minimal image. cheers holger
Hi holger could you add bug entry for the issues you think we should investigate? Stef On Wed, Jul 26, 2017 at 9:36 PM, Holger Freyther <holger@freyther.de> wrote:
On 22. Jul 2017, at 14:00, Hilaire <hilaire@drgeo.eu> wrote:
Hi Holger,
With Pharo3 I was proceeding like this, by uninstallation of packages. However there is the promise of building from a minimal image, but it is not documented AFAIK.
By curiosity, what is the size of your resulting image.
I see 31mb image size for Pharo3 and 49mb for Pharo6. The changes file is about the same. From the top command on a test system. "pharo" is a Pharo6 VM and image and pharo-vm is the pharo3 system.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 4206 root 20 0 1056684 74036 1600 S 5.7 1.2 1582:15 pharo-vm 10459 root 20 0 104040 90348 2124 S 5.0 1.5 369:00.29 pharo
I looked a bit at memory usage and see if some random changes make a difference
* Shrink ZnMimeType ExtensionMaps to a minimum when not handling extensions?
* ASTCache reset to get rid of the spotter cache. Not sure if the SessionManager runs ASTCache>>#shutDown properly. It looks like it should but it doesn't?
* ByteString compaction? I did strs := ByteString allInstances. str -> strs asSet. The numbers on a not fully clean image are 99874 vs. 42139. I wonder in which hell I end if I use >>#becomeForward: to compact it (I tried this once for GNU Smalltalk and I am sure others tried it before). I think anyone using == on String is using the wrong class? Any good arguments against doing it?
I am moving my Pharo6 image into production right now and then will look into setting up CI for Pharo7 with a minimal image.
cheers holger
participants (9)
-
Hilaire -
Holger Freyther -
phil＠highoctane.be -
Serge Stinckwich -
serge.stinckwich＠gmail.com
-
Stephane Ducasse -
Sven Van Caekenberghe -
Tim Mackinnon -
Tudor Girba