[tg-trunk] Re: TG2 release plan thoughts
On Thu, Dec 11, 2008 at 7:08 PM, Iain Duncan iaindun...@telus.net wrote: On Thu, 2008-12-11 at 15:21 -0600, Jorge Vargas wrote: On Wed, Dec 10, 2008 at 11:44 PM, Chris Miles miles.ch...@gmail.com wrote: On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. From the experience of someone that did that, You will be disappointed and you will learn a lot. From someone that comes from TG1 or django, pylons is like giving you a 1 pieces puzzle with no picture of the image and a lot of ocean and sky tiles. So you have to expend a lot of time reading about the components and understanding how things are done, and ones your finally done with all those blue tiles you have no energy to get started with the ship. And this is exactly the kind of thing TG2 is here for. You start right there at the ship and if you don't want to expend countless hours on the sky and sea, you don't have to, but if some part of it is important you can dive in there. OMG, that was hilarious Jorge. What a perfect metaphor. ;-) That's interesting as it just came out but given two people took the time to thank me I guess it's rather good :) I should blog about it or maybe write some inspiration documents based on the idea, I'll keep that in mind --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to turbogears-trunk+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
Of course Pylons knowledge doesn't harm. However, it is a little bit the other way around. I have now two webapps running TG2 and didn't use any Pylons before: TG2 is good jump into Pylons. It has a good quickstart template and chooses reasonable defaults. Setting some components as defaults also means, that the documentation can be more concrete: showing how to use problems with Genshi, SA, repoze.what in this combination. Of course, once you are more experienced, it is usually no problem to abstract from that description. So, I recommend you, just to start your next project with TG2 :-). Michael On 11 Dez., 06:44, Chris Miles [EMAIL PROTECTED] wrote: On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. For those that haven't already found it, James Gardner's Pylons book is available to read free online and has been invaluable for my Pylons learning. http://pylonsbook.com/ Cheers, Chris Miles --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
I totally agree with this. I had started using tg2 with only some background of tg1 and with no background of pylons behind me and I found that I could use tg2 to a large extent without problems. Fortunately when I had started on TG1 about 2 years back, I had started using tg1 along with SA as the ORM. Although a lots of TG goodies available with SO were not supported on SA at that time, I made the choice for SA as the devs had made their intentions of eventually migrating to SA very clear. So that was indeed a big plus for starting with TG2. But apart from that I would like to give full credits to Mark (and others) for making the TG2 controller api almost an exact replica of TG1. A bunch of wonderful new technologies have been employed in TG2 while keeping the api very similar. From what I gather, Gustavo, Florent, et al have made even the auth and auth api similar to that of tg1. I have worked with toscawidgets over the last one year and found it to be very similar to TG widgets. Yes, it had lacked some docs, but now the docs are quite decent. So while the knowledge of pylons is a definite plus towards fully understanding the framework, TG2 could be used even without knowing pylons. Another thing that I would like to mention here is that all the support and documentation provided on TG1 by ChrisA, Florent and Others has and will play a key role in overall success of TG2. Two things are worth mentioning here * Much of the TG1 wiki docs were used as starting point for TG2 docs * The support provided on TG1 has helped and will help TG1 users not to move away from TG and to stick to the framework and eventually move on to TG2 I agree with the suggestions of better marketing and an appealing template for TG2 as they help attract users. But I have another suggestion here. Can we have a quick 2nd edition of the TG book which includes WSGI concepts, TG2 api and TW, while people can keep working on a final good TG2 book in the long term. New users almost always look whether books are available. If not they just tend to move away. Finally, kudos to the entire TG team for this wonderful development and community experience :) Regards Sanjiv On Thu, Dec 11, 2008 at 5:05 PM, Michael Brickenstein [EMAIL PROTECTED] wrote: Of course Pylons knowledge doesn't harm. However, it is a little bit the other way around. I have now two webapps running TG2 and didn't use any Pylons before: TG2 is good jump into Pylons. It has a good quickstart template and chooses reasonable defaults. Setting some components as defaults also means, that the documentation can be more concrete: showing how to use problems with Genshi, SA, repoze.what in this combination. Of course, once you are more experienced, it is usually no problem to abstract from that description. So, I recommend you, just to start your next project with TG2 :-). Michael On 11 Dez., 06:44, Chris Miles [EMAIL PROTECTED] wrote: On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. For those that haven't already found it, James Gardner's Pylons book is available to read free online and has been invaluable for my Pylons learning. http://pylonsbook.com/ Cheers, Chris Miles --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Wed, Dec 10, 2008 at 11:44 PM, Chris Miles miles.ch...@gmail.com wrote: On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. From the experience of someone that did that, You will be disappointed and you will learn a lot. From someone that comes from TG1 or django, pylons is like giving you a 1 pieces puzzle with no picture of the image and a lot of ocean and sky tiles. So you have to expend a lot of time reading about the components and understanding how things are done, and ones your finally done with all those blue tiles you have no energy to get started with the ship. And this is exactly the kind of thing TG2 is here for. You start right there at the ship and if you don't want to expend countless hours on the sky and sea, you don't have to, but if some part of it is important you can dive in there. For those that haven't already found it, James Gardner's Pylons book is available to read free online and has been invaluable for my Pylons learning. http://pylonsbook.com/ --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to turbogears-trunk+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Thu, Dec 11, 2008 at 10:21 PM, Jorge Vargas jorge.var...@gmail.com wrote: On Wed, Dec 10, 2008 at 11:44 PM, Chris Miles miles.ch...@gmail.com wrote: On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. From the experience of someone that did that, You will be disappointed and you will learn a lot. From someone that comes from TG1 or django, pylons is like giving you a 1 pieces puzzle with no picture of the image and a lot of ocean and sky tiles. So you have to expend a lot of time reading about the components and understanding how things are done, and ones your finally done with all those blue tiles you have no energy to get started with the ship. And this is exactly the kind of thing TG2 is here for. You start right there at the ship and if you don't want to expend countless hours on the sky and sea, you don't have to, but if some part of it is important you can dive in there. Jorge, I like the metaphore. This is the differenciating tagline I had been searching for some time now: With tg2.0 begin by the ship not by the sky!. Joking :) You're quite right, it is interesting to note that TG2.0 is a good framework in the original sense of the word, that will help you work with pylons. And as you noted it is really easy to just do something the pylons way if you think tg is not doing this or that part as you would like it to be done. Florent. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to turbogears-trunk+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Thu, 2008-12-11 at 15:21 -0600, Jorge Vargas wrote: On Wed, Dec 10, 2008 at 11:44 PM, Chris Miles miles.ch...@gmail.com wrote: On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. From the experience of someone that did that, You will be disappointed and you will learn a lot. From someone that comes from TG1 or django, pylons is like giving you a 1 pieces puzzle with no picture of the image and a lot of ocean and sky tiles. So you have to expend a lot of time reading about the components and understanding how things are done, and ones your finally done with all those blue tiles you have no energy to get started with the ship. And this is exactly the kind of thing TG2 is here for. You start right there at the ship and if you don't want to expend countless hours on the sky and sea, you don't have to, but if some part of it is important you can dive in there. OMG, that was hilarious Jorge. What a perfect metaphor. ;-) Iain --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to turbogears-trunk+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Wed, Dec 10, 2008 at 2:59 AM, Iain Duncan [EMAIL PROTECTED] wrote: I assume this means rather than TG 1.9.7.beta3, we'll be seeing TG 2.0.0.beta1 next right? Seems about right, and that was the plan. Though I'm not opposed to keeping the numbering scheme we have now until the stable release becomes 2.0. I'm +1 on that. I think major release numbers create a lot of expectation in terms of polish and that it is a big marketing no-no to rush the major release numbers. I have the exact opposite opinion here: We have 1.0.x, 1.1, 1.5 and 1.9 branches... I strongly feel 2.0 should be identified as such. If we enter a cycle of beta versions we will then have a possible rc cycle. This means we have some margin before we label TG2 as stable. We need to remember TG2 was actually announced at least one year ago in great fanfare, the evil is already done and I prefer having a 2.0beta with some quirks than people calling tg2 vaporware. At the moment tg2 _is_ vaporware for the vast majority of python users out there. I have been in a lot of python enthusiast meetings, the people I met there were surprised that TG2 actually existed! TG2 does not exist, TG2 does not have a presence, if we continue to hide TG2 behind a 1.9.xxbetax versioning scheme we're just finding excuses to not release the damn thing and to fail silently. I have invested time and effort in TG1 to maintain the platform waiting for TG2 to mature, I have invested time and effort to help TG2 get closer to a release when/where I could, but I feel now is the good moment to start going public and to start pushing the thing out. If we don't do it now we won't do it ever. Remember that open source developers come and go and the more you speak and go public about your product the more developers come, the more you go shy and silent about your product the more developers go elsewhere. I feel our platform is stable and robust enough to justify a beta cycle (I'm not talking about a stable shiny 2.0 here). Florent. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Dec 9, 12:58 am, Mark Ramm [EMAIL PROTECTED] wrote: TurboGears 2 is somewhat uneque in the modern web framework space in that it attempts to provide solutions for the archetectual problems that limited sites like Pownce, Twitter, and TG1 sites, that I've worked on in the past. I'm very interested to hear about how TG2 will avoid the scaling and traffic issues that hit pownce and twitter. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
I'm very interested to hear about how TG2 will avoid the scaling and traffic issues that hit pownce and twitter. Well, there are lots of issues, and without being super-connected to either Twitter or Pownce, one of the things that seems to have happened to both early on is that it was difficult for their platform of choice to connect to multiple databases and handle horizontal partitioning. Beyond that, it was hard for them to manage how many DB hits each page produced since the ORM's in their platform didn't make it easy to group sets of db changed together and submit them all at once. Nor did they have the flexibility that SQLAlchemy provides in grouping which data will be returned by a request of a particular mapped object, both because SA provides fine-grained control of what will be eager-loaded and what should be lazily loaded, and because SA as a Data Mapper implementation allows objects to map to data from several tables at once. Of course both Pownce and Twitter eventually got to the point where they were using hand-coded SQL and multiple databases for lots of things, and were still having database scalability issues. And there's no magic that the framework can do to fix that. But SQLAlchemy continues to shine as a toolkit even when you're not using the ORM. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Wed, 2008-12-10 at 09:47 -0500, Mark Ramm wrote: I'm very interested to hear about how TG2 will avoid the scaling and traffic issues that hit pownce and twitter. Well, there are lots of issues, and without being super-connected to either Twitter or Pownce, one of the things that seems to have happened to both early on is that it was difficult for their platform of choice to connect to multiple databases and handle horizontal partitioning. Beyond that, it was hard for them to manage how many DB hits each page produced since the ORM's in their platform didn't make it easy to group sets of db changed together and submit them all at once. Nor did they have the flexibility that SQLAlchemy provides in grouping which data will be returned by a request of a particular mapped object, both because SA provides fine-grained control of what will be eager-loaded and what should be lazily loaded, and because SA as a Data Mapper implementation allows objects to map to data from several tables at once. You know, a detailed tutorial/article on the above would probably be very good TG2 publicity, as long as you can do so without looking like you're using Twitter/Pownce as scapegoats. Iain --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
Timur Izhbulatov schrieb: I feel that my impressions of learning TG 2 are getting stale. I'm going to share some random notes with you (and everybody else) before it's to late... :) Sorry for long email. Thanks for writing this up. So it seems we will need a everything a TG2 user needs to know about WSGI article as well ;-) I hope I can find some time to have a look at TG2 soon, and contribute something like that or otherwise improve the TG2 docs. -- Chris --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On 11/12/2008, at 1:47 AM, Mark Ramm wrote: I'm very interested to hear about how TG2 will avoid the scaling and traffic issues that hit pownce and twitter. Well, there are lots of issues, and without being super-connected to either Twitter or Pownce, one of the things that seems to have happened to both early on is that it was difficult for their platform of choice to connect to multiple databases and handle horizontal partitioning. Various levels of caching are also important for scaling high traffic sites. Does TG2 offer any inbuilt object and/or template caching mechanism (using memcached or similar)? Or does it rely on Pylons/Beaker for caching? While learning some Django recently I was pleased to see that it offers caching support out of the box. Pylons offers caching with Beaker. I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Cheers, Chris Miles --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. --Mark Ramm --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On 11/12/2008, at 4:36 PM, Mark Ramm wrote: I assume Beaker is also the recommended caching solution for TG2, but would like clarification. Yes, beaker is the built in caching mechanism of TG2. Beaker supports memcached, and lots of other back-ends. It also doesn't suffer from the so-called dog-pile effect i the same way as some other web-framework's built in caching mechanisms do, because it's just plain awesome. That's good. I've decided to learn standard Pylons before looking at TG2 in detail. I assume a lot of Pylons knowledge will make working with TG2 much easier. For those that haven't already found it, James Gardner's Pylons book is available to read free online and has been invaluable for my Pylons learning. http://pylonsbook.com/ Cheers, Chris Miles --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
Mark, As a user, I tend to think of using TurboGears as not only a framework, but kind of like a full out distribution. I prefer to use the packages you guys include in your own repo, rather than pull individual components myself to make sure it's all as stable as possible. I'd love to see a final/stable release for 2.0, however I would ask you to please hold off at least until all your dependencies have their own stable releases. Most importantly SQLAlchemy and Pylons, which are both in beta/RC phases still. I'm sure we're still a few months away from TG2.0.final anyway, so hopefully this won't even be an issue by then. Thanks to all of you, it has been a pleasure working with the alphas and betas so far ;) - Derick Eisenhardt --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
I'd love to see a final/stable release for 2.0, however I would ask you to please hold off at least until all your dependencies have their own stable releases. Most importantly SQLAlchemy and Pylons, which are both in beta/RC phases still. I'm sure we're still a few months away from TG2.0.final anyway, so hopefully this won't even be an issue by then. I actually think both SA and Pylons are in the RC process now, so there should be no API changes, only bugfixes between now and stable releases. So I don't think we should stop trying to get TG2 ready for final to wait for them... But you're right, if we can't get stable pylons and SA releases, we may need to hold off on the final TG2 release for that. --Mark Ramm --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
Mark Ramm schrieb: If you're interested in joining this support team, please let us know. If you think you can help get it organized and make sure that it runs smoothly defintely let us know. It's good to see TG2 making progress. So far I've focused on TG 1.x which I'm using for my productive apps, but I'd also like to work with/on TG 2 in the future. I think the main reason why I didn't really look into TG 2 so far was that I never used Pylons, so I feel I first need to get aquainted with Pylons, but haven't found the time so far. It would be nice to have a kind of Pylons crash course available, or a everything a TG user needs to know about Pylons article. Maybe something like this already exist? Can you give a good starting point and roadmap for learning Pylons and TG 2? -- Christoph --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
I think the main reason why I didn't really look into TG 2 so far was that I never used Pylons, so I feel I first need to get aquainted with Pylons, but haven't found the time so far. It would be nice to have a kind of Pylons crash course available, or a everything a TG user needs to know about Pylons article. This is a great idea, and I'll write something like this this week sometime. --Mark Ramm --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
One more quick thought... On Dec 8, 11:58 pm, Mark Ramm [EMAIL PROTECTED] wrote: I'd like to propose that we declare the next stable TG2 release 2.0 and that we work together to get a release candidate by the end of the year. I assume this means rather than TG 1.9.7.beta3, we'll be seeing TG 2.0.0.beta1 next right? --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Dec 8, 11:58 pm, Mark Ramm [EMAIL PROTECTED] wrote: I'd like to propose that we declare the next stable TG2 release 2.0 and that we work together to get a release candidate by the end of the year. I assume this means rather than TG 1.9.7.beta3, we'll be seeing TG 2.0.0.beta1 next right? Seems about right, and that was the plan. Though I'm not opposed to keeping the numbering scheme we have now until the stable release becomes 2.0. --Mark Ramm --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Tue, Dec 9, 2008 at 5:45 PM, Mark Ramm [EMAIL PROTECTED] wrote: On Dec 8, 11:58 pm, Mark Ramm [EMAIL PROTECTED] wrote: I'd like to propose that we declare the next stable TG2 release 2.0 and that we work together to get a release candidate by the end of the year. I assume this means rather than TG 1.9.7.beta3, we'll be seeing TG 2.0.0.beta1 next right? Seems about right, and that was the plan. Though I'm not opposed to keeping the numbering scheme we have now until the stable release becomes 2.0. I think the 2.0 is a great idea but I would put some pressure on Pylons to release the 1.0 version. They seem to be taking forever on the 0.9.6. I'm not sure but the last time I've checked (6 months ago ) the only thing that needed to be done was docs. It seemed then it will take max a month to make a 1.0 release. It seems to me they are holding back. I'm not sure if there are any developers here that are also active on pylons and can talk to people in control to release a 1.0 and work on 1.0.1. It seems to me that pylons won't be changing much in a next 6 months and they are just floating. I think making a 1.0 release would give a boost to them and make the tg2 available! Can we start a discussion on pylons devel about that? Based on their roadmap (http://pylonshq.com/project/) there doesn't seem to be any major bugs to incompatibilities. That is my impression. If they say they won't release 1.0 until routes 2 , new webhelprs and pylonsbook comes out and if it will take more then 2 months then we should release a turbogears 2.0. As far as sqlalchemy, they are on a strict schedule of releases and after they went into 0.5 it seems as they will continue on a steady path. It seems to me that 0.5 is as stable as it gets, we should not wait just so we can see a different number. It was stable year ago and it is stable now. Thanks. Lucas --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
2008/12/10 Christoph Zwerschke [EMAIL PROTECTED]:
Mark Ramm schrieb:
If you're interested in joining this support team, please let us know.
If you think you can help get it organized and make sure that it
runs smoothly defintely let us know.
It's good to see TG2 making progress. So far I've focused on TG 1.x
which I'm using for my productive apps, but I'd also like to work
with/on TG 2 in the future. I think the main reason why I didn't really
look into TG 2 so far was that I never used Pylons, so I feel I first
need to get aquainted with Pylons, but haven't found the time so far. It
would be nice to have a kind of Pylons crash course available, or a
everything a TG user needs to know about Pylons article. Maybe
something like this already exist? Can you give a good starting point
and roadmap for learning Pylons and TG 2?
Christoph,
I feel that my impressions of learning TG 2 are getting stale. I'm
going to share some random notes with you (and everybody else) before
it's to late... :) Sorry for long email.
Like you, I had been already using TG 1 for a while when I decided to
give TG 2 a try. This September I started working on a simple TG 2
application. It needs neither auth{entication,orization} nor database,
so I'm not discussing these here.
I would recommend starting from the official docs. In particular,
Getting Started with TurboGears and Tutorials were very helpful. There
were some example code snippets not working as is, but it was pretty
easy to figure out the problem and, probably, some or all of them are
fixed now.
Get familiar with WSGI if you aren't already. TG 2 was my first WSGI
framework. I tried looking at WSGI maybe few years ago. I read the PEP
(OK, everything looks pretty simple), took a look at pythonpaste.org
(what is this all about?!) ... and failed to get started with this
technology. But thanks to TG 2 now I have an idea how WSIG can be used
in practice.
Another good thing is that TG 2 has very small and code base written
in decent PEP 8 Python! I would recommend reading configuration.py
[1] first. This is the heart of the framework where all WSGI set up
happens. This module gives you a good idea of how TG 2 uses all
third-party components, and therefore how you can customize this in
your app. Other notable parts are controllers.py [2] and decorators.py
[3] telling you more details about object dispatch, validation,
rendering and some i18n.
Surprisingly, as of my impression TG 2 is not that much of Pylons from
application developer's perspective. TG 2 is more like a higher level
built *on top* of Pylons, so some concepts/recipes from the later are
not directly usable (decorators, for instance). And without looking at
the source code it's not always clear which things should be imported
from tg and which from pylons or somewhere else.
Decorators are *very* different from TG 1. Actually, now they are much
cleaner and use no magic — it's just hooks. To my knowledge, there's
no any examples in the official docs yet but I have some in my app
[4].
i18n tools are better and easier to use than in TG 1. However, in
general i18n still needs some workarounds [5].
I think the most difficult part of the learning process was
ToscaWidgets. This is a very powerful but complex framework, and when
you start using it, you quickly exhaust the material available on the
official site and in TG 2 docs. However, the source code is also
rather clean and reading some parts of it helped me a lot.
Best Regards,
Timur
[1] http://trac.turbogears.org/browser/trunk/tg/configuration.py
[2] http://trac.turbogears.org/browser/trunk/tg/controllers.py
[3] http://trac.turbogears.org/browser/trunk/tg/decorators.py
[4]
http://code.google.com/p/posy/source/browse/trunk/posy.tg/posy/tg/lib/decorators.py#23
[5] http://groups.google.com/group/turbogears-trunk/msg/21e0e32f43ea4e25
--
Timur Izhbulatov -- www.timka.org
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
TurboGears Trunk group.
To post to this group, send email to turbogears-trunk@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/turbogears-trunk?hl=en
-~--~~~~--~~--~--~---
[tg-trunk] Re: TG2 release plan thoughts
On Tue, 2008-12-09 at 18:45 -0500, Mark Ramm wrote: On Dec 8, 11:58 pm, Mark Ramm [EMAIL PROTECTED] wrote: I'd like to propose that we declare the next stable TG2 release 2.0 and that we work together to get a release candidate by the end of the year. I assume this means rather than TG 1.9.7.beta3, we'll be seeing TG 2.0.0.beta1 next right? Seems about right, and that was the plan. Though I'm not opposed to keeping the numbering scheme we have now until the stable release becomes 2.0. I'm +1 on that. I think major release numbers create a lot of expectation in terms of polish and that it is a big marketing no-no to rush the major release numbers. my two cents Iain --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups TurboGears Trunk group. To post to this group, send email to turbogears-trunk@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en -~--~~~~--~~--~--~---
