November 19, 2012 at 11:18Start of OpenSource Project
I think each of you at least once came across the question of choosing a project management system for your needs or the needs of a development group. And probably had to try several different systems before settling on a more or less suitable, and sometimes instead they sat down to write their own bike. So I also ran into this problem and not even more than once. It always so happened that I was the initiator and organizer of the teams for the development of several medium-sized projects. And I was responsible for organizing the working environment, including the SOUP. But I couldn’t find a solution completely suitable for me.
All the moss of trial and error in this direction will be enough for a separate article, but the purpose of this topic is completely different. Therefore, I will only briefly describe some of what I tried and what specifically I didn’t like, so that you better understand what I'm talking about.
1. Redmine. The system, excellent in functionality, works quickly, has many modules, a nice interface, and is free. The undisputed leader. I did not like it: the complexity of installation, migration and updating. I'm just a php programmer and the administration skills are very mediocre, which will fit the LAMP or MySQL settings. But the Ruby application turned out to be complicated for me. In debian, the apt-get install utility solves many problems, but the distributions are old. Install from source - I would not want to. In general, I had to solve many problems and lose a lot of time before redmine was installed and put into working condition.
2. Trac. Nice system. There is everything you need for a small team working on one or more projects. But the same problem with installation and transfers. Of course rtfm, google and some mother helped solve all the problems, but again a waste of time and installing additional packages on the server.
3. Mantis. This is a bug tracker, as everyone knows, but it can be used simply as a task manager. I didn’t like it: disgusting appearance, already my eyes get tired, the porridge inside (I wanted to finish it a bit for myself, since it is written in php), to change the skin you need to rewrite almost the entire product, there is no wiki.
4. JIRA. I can’t say anything bad, except that the system is paid and, again, this is primarily a bug tracker.
5. TeamLab. In the SaaS version it works very slowly, sometimes it falls off without warning. Still more sharpened by CRM. In the case of deploying "at home" you need expensive windows hosting.
Next came Megaplan and Bitrix, about which nobody needs to say anything, and so everyone knows everything, and many more systems that he did not even try.
As a result, for all the systems that I tried there were basically the same problems: either the complexity of installation and porting, or insufficient functionality, or paid, or only the SaaS version.
How do I see the system for myself:
1. php + mysql (or another database supported by PDO).
2. Installation is reduced to a simple run of Install.php in an archive previously unpacked on the hosting.
3. Transfer is possible stupidly by copying, ideally a built-in utility that first makes an archive of files and databases, and then deploys it in a new place (in one CMS, whose name is not called out loud, there is such a very convenient system).
4. ease of development and implementation of skins.
5. high speed of work
6. tasks, projects, task categories, version of projects, wiki, timeline, etc. from the basic necessary set.
7. api for writing extensions.
8. is free.
Thus, we get a system that can be easily and quickly deployed almost anywhere by almost any php programmer and will provide the necessary functionality for work. So that at the beginning of work on the next project to think only about the project, and not on organizational issues. Unfortunately, there is no such system at the moment.
And I was about to sit down to write my bike, but then I thought: but this will really be the next bike, which is likely to remain in my walls and except for me it will not bring anyone else any benefit and will not solve all those problems that I encountered I, and probably will die before reaching the finish line stupidly due to lack of time. In addition, for sure, during the development, I will involuntarily make amendments to my needs and from the heap of ideas only the most necessary for me will be implemented.
Therefore, I decided to start that there will be much more from an open source project. I must say right away that at the moment this is just an idea and no more. Before starting to do something, I sent several invitations to my friends and acquaintances to participate, and now I turn to the habrasociety with two questions:
1. Does anyone else need such a project besides me?
2. Are there any among the hawkers who would like to participate?
If it turns out that such a project is really needed and there are a few who want to participate, then I have the organization of work (repositories, hosting, domain, etc.) and also full participation in the development.
Z.Y. The writer of articles from me unimportant, unfortunately, outlined as he could.
UPD: If there are those who wish, then I will organize a collective discussion of the details of the project.
All the moss of trial and error in this direction will be enough for a separate article, but the purpose of this topic is completely different. Therefore, I will only briefly describe some of what I tried and what specifically I didn’t like, so that you better understand what I'm talking about.
1. Redmine. The system, excellent in functionality, works quickly, has many modules, a nice interface, and is free. The undisputed leader. I did not like it: the complexity of installation, migration and updating. I'm just a php programmer and the administration skills are very mediocre, which will fit the LAMP or MySQL settings. But the Ruby application turned out to be complicated for me. In debian, the apt-get install utility solves many problems, but the distributions are old. Install from source - I would not want to. In general, I had to solve many problems and lose a lot of time before redmine was installed and put into working condition.
2. Trac. Nice system. There is everything you need for a small team working on one or more projects. But the same problem with installation and transfers. Of course rtfm, google and some mother helped solve all the problems, but again a waste of time and installing additional packages on the server.
3. Mantis. This is a bug tracker, as everyone knows, but it can be used simply as a task manager. I didn’t like it: disgusting appearance, already my eyes get tired, the porridge inside (I wanted to finish it a bit for myself, since it is written in php), to change the skin you need to rewrite almost the entire product, there is no wiki.
4. JIRA. I can’t say anything bad, except that the system is paid and, again, this is primarily a bug tracker.
5. TeamLab. In the SaaS version it works very slowly, sometimes it falls off without warning. Still more sharpened by CRM. In the case of deploying "at home" you need expensive windows hosting.
Next came Megaplan and Bitrix, about which nobody needs to say anything, and so everyone knows everything, and many more systems that he did not even try.
As a result, for all the systems that I tried there were basically the same problems: either the complexity of installation and porting, or insufficient functionality, or paid, or only the SaaS version.
How do I see the system for myself:
1. php + mysql (or another database supported by PDO).
2. Installation is reduced to a simple run of Install.php in an archive previously unpacked on the hosting.
3. Transfer is possible stupidly by copying, ideally a built-in utility that first makes an archive of files and databases, and then deploys it in a new place (in one CMS, whose name is not called out loud, there is such a very convenient system).
4. ease of development and implementation of skins.
5. high speed of work
6. tasks, projects, task categories, version of projects, wiki, timeline, etc. from the basic necessary set.
7. api for writing extensions.
8. is free.
Thus, we get a system that can be easily and quickly deployed almost anywhere by almost any php programmer and will provide the necessary functionality for work. So that at the beginning of work on the next project to think only about the project, and not on organizational issues. Unfortunately, there is no such system at the moment.
And I was about to sit down to write my bike, but then I thought: but this will really be the next bike, which is likely to remain in my walls and except for me it will not bring anyone else any benefit and will not solve all those problems that I encountered I, and probably will die before reaching the finish line stupidly due to lack of time. In addition, for sure, during the development, I will involuntarily make amendments to my needs and from the heap of ideas only the most necessary for me will be implemented.
Therefore, I decided to start that there will be much more from an open source project. I must say right away that at the moment this is just an idea and no more. Before starting to do something, I sent several invitations to my friends and acquaintances to participate, and now I turn to the habrasociety with two questions:
1. Does anyone else need such a project besides me?
2. Are there any among the hawkers who would like to participate?
If it turns out that such a project is really needed and there are a few who want to participate, then I have the organization of work (repositories, hosting, domain, etc.) and also full participation in the development.
Z.Y. The writer of articles from me unimportant, unfortunately, outlined as he could.
UPD: If there are those who wish, then I will organize a collective discussion of the details of the project.