I honestly don't mean to sound ungrateful (especially seeing that the product is free), but for the life of me, I can't figure out how this entire project has gone so far off the rails and what the "core" team is doing to fix it.
I mean honestly folks, the whole "5" disaster was predictable. There are too many things going on at once and yet nothing important getting fixed. I have kept an eye on codeplex and these forums for quite some time and see the same old bugs OVER and OVER and NOBODY fixing them or even assigning them, yet we keep getting NEW releases with NEW features. I see the same old confusing (or lack of or contradictory) documentation and it never really gets better.
I honestly would have thought that as bugs appeared in codeplex that core team members would derive fixes for them as a prioroty over new features or code branches.
To that end, I am also finding OLD bugs that were fixed in previous releases present again in new releases. I just updated a site from 4.5 to 4.7.2 and found the good old "medier" bug is back. Why? There really is no excuse for this.
The spellchecker does not work... why? There is no excuse for this?
The upgrade instructions are STILL TERRIBLE. One has to visit a dozen forum posts and 3rd party internet sites to figure out what files need to be removed from the BIN folder. There is NO EXCUSE for this.
Bugs: The entire concept of "voting" bugs up is silly at best but even at that, most of them (regardless of votes or severity) get outright ignored. Again, I would think that the list would be actively worked, but it is not.
Part of the new promise when "5" was canceled was to refocus on 4.7 and get it right. I think many of us took that at face value and assumed that 5 would be dropped in favor of finishing 4.7, not starting Umbraco WPF 1.0 and adding more bugs in the form of 4.8 while still ignoring the PILE of legacy bugs STILL In 4.7.
Again, not to sound brash (okay maybe I mean to) what does it take to get somebody to run this project in an organized manner and work through the bugs that exist, and have existed for a year or years? What does it take to get a clear concise upgrade document that is KEPT up to date for each version. I dread the Umbraco upgrade process, as it ALWAYS involves hours and/or days of reading and low level file manipulation. There has to be a better way.
Since Umbraco is an open source project it heavily relies on contributions to the core from the community. Yes I know Umbraco HQ now have about 11-12 people on the pay list but the project has grown so much during the past couple of years that not all of them focus on developing Umbraco.
Since the primary focus has been on v5 for the past 1.5-2 years or so it's only natural that the old version does not get as much love as it might should have gotten. It's probably because people have been waiting for v5 being ready for production and even though they knew how to fix bugs on v4 why should they bother? V5 was supposed to be ready soon...I think that's why not many bugs have been fixed the past year since people don't want to waste their time fixing something that was about to be a legacy version of Umbraco. So I think it's only natural that several v4 bugs has not been fixed. Focus was somewhere else.
Unfortunately along the way in the process of developing v5 the connection between umbraco hq and the community somehow got lost. V5 became a project driven by HQ with not that many core contributors and the community was excited about this new v5 stuff and was busy getting up to speed with MVC, Hive etc.- If we as a community had raised our voices about all the issues that many of us have experienced during hackathons, courses and trying to do projects on v5 maybe this situation could have been avoided or maybe v5 would have been killed even earlier on. No matter what it's frustrating for many people that it has happened it seems that many people are also relieved that v5 has been killed and focus is back on v4 (I'm one of them). I've seen much more activity on this forum, twitter and on the mailing list on google groups than I have seen for a long time. The community is breathing again!
I think many of us can relate to your frustrations about bugs that had been fixed in a previous version re-appering in a new version. That is really annyoing and it's hard to remember, in what version it works if you're dealing with clients running a wide range of Umbraco versions. I'm always carefull when demoing the "publish at" feature to a client since I can't remember in, which versions it's actually working as expected.
Why bugs reappear I have no clue and obviously it's something that should be fixed so it does not happen again. One of the initiatives that I believe will be used to avoid this is to have new releases tested more thouroughly before they go public. I think this is also becaue of the code base of v4 being so bloated that it's hard to maintain - fortunately this is something that is going to get fixed over time so the codebase becomes cleaner and easier to deal with for everyone.
You do have a point about the voting system not being the ideal solution anymore and if I remember correctly the way bugs are being prioritized will change in the future - And if you have some ideas about how to handle it more efficient I think that the whole community are more than interested to listen :)
HQ are going to pay more attention to do project managing of Umbraco so it will be handled in a more organized manner in the future. But do keep in mind that focus has been on v5 and therefore v4 has not gotten that much attention. There is a pile of bug reports on codeplex with bugs that may or may not be valid anymore so there is a bit of a cleanup to be done before anyone has an overview of the most critical things to fix.
If you have the skills and some spare time and you're willing to help I think now is the right time to contribute to the Umbraco project - It does not neccesarily need to be on fixing the Umbraco core it could also be documenting how stuff in Umbraco works, since the documentation project has now been focused on v4 instead of v5.
Thanks for sharing your frustrations - I hope it has given you some space to breathe and that you at some point is able to look forward and help bringing Umbraco to the state where we all want it to be: The best, most friendly and fun CMS to work with :)
Thank you for taking the time to respond. I am not sure if there will continue to be a "paid" core, or the bulk of the development will again fall back to volunteer coders. In either case, my suggestion would be to designate a person or team to direct the workflow and ensure that bugs get the attention they need.
As an early adopter of Umbraco, it is my feeling that the "problem" has been systemic in the workflow since the onset. We have always had the allure of new versions and features outweighing the "old version" and proper documentation. In that sense, bugs have always piled up and documentation always lagged behind in favor of a new release with the latest acronym technology and featureset. There is no blame to assign, it is the nature of such projects, we just need to find a way to control it.
With the reality of the past in mind and as there is already a mess (bloated code, recurring bugs, and no specific direction for many of the coders, etc) I would assert that an immediate feature freeze be put in place until 4.7.2 is fully sorted out and ALL. new and old bugs addressed (maybe not fixed, but addressed and updated in codeplex). At that point work can begin on fixes and a roadmap for future milestones be addressed. In that spirit, part of eachmilestone criteria would be ensuring complete bug fixes, code optimization and documentation for the current level of development before ANYBODY begins work on the next milestone. I would think that such an approach would be more efficient with regard to both paid and volunteer time and the end result will be a more stable and better documented CMS that IS truley fun and friendly.
Such a framework would require a project manager or team that is willing to stick to the framework and drive focus for each milestone and ensure the criteria are met for each milestone.
I would be more than happy to contribute to the core, but I fear my coding skills were barely capable in VB6, back when things were simple. I have attempted to contribute documentation but have found it almost overwhelming due to my lack of understanding of the source.
Who is asleep at the wheel?
I honestly don't mean to sound ungrateful (especially seeing that the product is free), but for the life of me, I can't figure out how this entire project has gone so far off the rails and what the "core" team is doing to fix it.
I mean honestly folks, the whole "5" disaster was predictable. There are too many things going on at once and yet nothing important getting fixed. I have kept an eye on codeplex and these forums for quite some time and see the same old bugs OVER and OVER and NOBODY fixing them or even assigning them, yet we keep getting NEW releases with NEW features. I see the same old confusing (or lack of or contradictory) documentation and it never really gets better.
I honestly would have thought that as bugs appeared in codeplex that core team members would derive fixes for them as a prioroty over new features or code branches.
To that end, I am also finding OLD bugs that were fixed in previous releases present again in new releases. I just updated a site from 4.5 to 4.7.2 and found the good old "medier" bug is back. Why? There really is no excuse for this.
The spellchecker does not work... why? There is no excuse for this?
The upgrade instructions are STILL TERRIBLE. One has to visit a dozen forum posts and 3rd party internet sites to figure out what files need to be removed from the BIN folder. There is NO EXCUSE for this.
Bugs: The entire concept of "voting" bugs up is silly at best but even at that, most of them (regardless of votes or severity) get outright ignored. Again, I would think that the list would be actively worked, but it is not.
Part of the new promise when "5" was canceled was to refocus on 4.7 and get it right. I think many of us took that at face value and assumed that 5 would be dropped in favor of finishing 4.7, not starting Umbraco WPF 1.0 and adding more bugs in the form of 4.8 while still ignoring the PILE of legacy bugs STILL In 4.7.
Again, not to sound brash (okay maybe I mean to) what does it take to get somebody to run this project in an organized manner and work through the bugs that exist, and have existed for a year or years? What does it take to get a clear concise upgrade document that is KEPT up to date for each version. I dread the Umbraco upgrade process, as it ALWAYS involves hours and/or days of reading and low level file manipulation. There has to be a better way.
Hi William
Since Umbraco is an open source project it heavily relies on contributions to the core from the community. Yes I know Umbraco HQ now have about 11-12 people on the pay list but the project has grown so much during the past couple of years that not all of them focus on developing Umbraco.
Since the primary focus has been on v5 for the past 1.5-2 years or so it's only natural that the old version does not get as much love as it might should have gotten. It's probably because people have been waiting for v5 being ready for production and even though they knew how to fix bugs on v4 why should they bother? V5 was supposed to be ready soon...I think that's why not many bugs have been fixed the past year since people don't want to waste their time fixing something that was about to be a legacy version of Umbraco. So I think it's only natural that several v4 bugs has not been fixed. Focus was somewhere else.
Unfortunately along the way in the process of developing v5 the connection between umbraco hq and the community somehow got lost. V5 became a project driven by HQ with not that many core contributors and the community was excited about this new v5 stuff and was busy getting up to speed with MVC, Hive etc.- If we as a community had raised our voices about all the issues that many of us have experienced during hackathons, courses and trying to do projects on v5 maybe this situation could have been avoided or maybe v5 would have been killed even earlier on. No matter what it's frustrating for many people that it has happened it seems that many people are also relieved that v5 has been killed and focus is back on v4 (I'm one of them). I've seen much more activity on this forum, twitter and on the mailing list on google groups than I have seen for a long time. The community is breathing again!
I think many of us can relate to your frustrations about bugs that had been fixed in a previous version re-appering in a new version. That is really annyoing and it's hard to remember, in what version it works if you're dealing with clients running a wide range of Umbraco versions. I'm always carefull when demoing the "publish at" feature to a client since I can't remember in, which versions it's actually working as expected.
Why bugs reappear I have no clue and obviously it's something that should be fixed so it does not happen again. One of the initiatives that I believe will be used to avoid this is to have new releases tested more thouroughly before they go public. I think this is also becaue of the code base of v4 being so bloated that it's hard to maintain - fortunately this is something that is going to get fixed over time so the codebase becomes cleaner and easier to deal with for everyone.
You do have a point about the voting system not being the ideal solution anymore and if I remember correctly the way bugs are being prioritized will change in the future - And if you have some ideas about how to handle it more efficient I think that the whole community are more than interested to listen :)
HQ are going to pay more attention to do project managing of Umbraco so it will be handled in a more organized manner in the future. But do keep in mind that focus has been on v5 and therefore v4 has not gotten that much attention. There is a pile of bug reports on codeplex with bugs that may or may not be valid anymore so there is a bit of a cleanup to be done before anyone has an overview of the most critical things to fix.
If you have the skills and some spare time and you're willing to help I think now is the right time to contribute to the Umbraco project - It does not neccesarily need to be on fixing the Umbraco core it could also be documenting how stuff in Umbraco works, since the documentation project has now been focused on v4 instead of v5.
Thanks for sharing your frustrations - I hope it has given you some space to breathe and that you at some point is able to look forward and help bringing Umbraco to the state where we all want it to be: The best, most friendly and fun CMS to work with :)
Cheers,
Jan
Jan,
Thank you for taking the time to respond. I am not sure if there will continue to be a "paid" core, or the bulk of the development will again fall back to volunteer coders. In either case, my suggestion would be to designate a person or team to direct the workflow and ensure that bugs get the attention they need.
As an early adopter of Umbraco, it is my feeling that the "problem" has been systemic in the workflow since the onset. We have always had the allure of new versions and features outweighing the "old version" and proper documentation. In that sense, bugs have always piled up and documentation always lagged behind in favor of a new release with the latest acronym technology and featureset. There is no blame to assign, it is the nature of such projects, we just need to find a way to control it.
With the reality of the past in mind and as there is already a mess (bloated code, recurring bugs, and no specific direction for many of the coders, etc) I would assert that an immediate feature freeze be put in place until 4.7.2 is fully sorted out and ALL. new and old bugs addressed (maybe not fixed, but addressed and updated in codeplex). At that point work can begin on fixes and a roadmap for future milestones be addressed. In that spirit, part of eachmilestone criteria would be ensuring complete bug fixes, code optimization and documentation for the current level of development before ANYBODY begins work on the next milestone. I would think that such an approach would be more efficient with regard to both paid and volunteer time and the end result will be a more stable and better documented CMS that IS truley fun and friendly.
Such a framework would require a project manager or team that is willing to stick to the framework and drive focus for each milestone and ensure the criteria are met for each milestone.
I would be more than happy to contribute to the core, but I fear my coding skills were barely capable in VB6, back when things were simple. I have attempted to contribute documentation but have found it almost overwhelming due to my lack of understanding of the source.
Thank you again from the prompt reply!
Bill
is working on a reply...