First off I want to preface what I am about to say with the fact that I have been using moodle for over 3 years and sign its praises every chance I get. The staff where I teach love it. The students and parents are really coming on board. Those that do the "grunt" work on moodle are fantastic and I truly admire you and your efforts!!
I've upgrade many times from about 1.1 now to 1.7 and things have always gone really well (unless I do something dumb - known to happen). 1.7's upgrade when flawless... but things have never been the same.
I've been having many issues:
http://moodle.org/mod/forum/discuss.php?d=58674
http://moodle.org/mod/forum/discuss.php?d=58548
http://moodle.org/mod/forum/discuss.php?d=58489
http://moodle.org/mod/forum/discuss.php?d=58673
http://moodle.org/mod/forum/discuss.php?d=58497
I'm considering rolling back to 1.6.3 but would like to know what others are experience with 1.7?
Jeff
PS - this isn't meant as a "spleen vent" The scientist in me is looking to collect data.
The data is all in http://tracker.moodle.org already, Jeff, assuming people have been filing issues as requested.
Some of these seem to be unfamiliarity with Roles (understandable because they can be complex!). Others of these seem to be real issues that did not surface during the Beta test period - there's only so much a small team of core developers can do .. we depend on testers and contributors to find and report the more obscure bugs and help us get them fixed. Roles is a radical change to Moodle that affected pretty much every single script in it.
My experience for Moodle 1.7 was that only a very small group of people really engaged in the testing process, despite our rally cry being louder than ever (I thought). This is something we need to improve on as a community if we want to see fewer bugs in releases.
Some of these seem to be unfamiliarity with Roles (understandable because they can be complex!). Others of these seem to be real issues that did not surface during the Beta test period - there's only so much a small team of core developers can do .. we depend on testers and contributors to find and report the more obscure bugs and help us get them fixed. Roles is a radical change to Moodle that affected pretty much every single script in it.
My experience for Moodle 1.7 was that only a very small group of people really engaged in the testing process, despite our rally cry being louder than ever (I thought). This is something we need to improve on as a community if we want to see fewer bugs in releases.
Thanks for your prompt reply Martin. I truly appreciate all that you and your core programmers do!
It would appear that my failure (and others) to test 1.7 beta in a large environment has lead to some of problems. I always play with the beta versions, but only have a few "test" accounts I use and didn't notice the issues.
I really wish I could program so i could offer my assistance, but unfortunately my programming using BASIC won't help
Jeff
It would appear that my failure (and others) to test 1.7 beta in a large environment has lead to some of problems. I always play with the beta versions, but only have a few "test" accounts I use and didn't notice the issues.
I really wish I could program so i could offer my assistance, but unfortunately my programming using BASIC won't help
Jeff
Just throwing in my humble 2 cents. . .
I would think that release timing is part of the problem. Certainly with such a globally used application there will NEVER be a perfect time to release, but at least in the US we are very locked down to our school calendars (isn't everyone? the calendars are all just a little different). I would think that few IT departments in large installations would be very willing to upgrade in the middle of a semester, or even in the middle of an academic year. That means your testing pool is smaller.
From our perspective, 1.6 came out in August (wasn't it?). I went ahead with it because of the essay questions in the quizzes. That was important to us. But we use several 3rd-party blocks and modules. Some of which were not too hard to get into operational order (with a little help from our friends here on the forums!) and some of which we have given up on for now (upload and review, for example). We were struggling to get that up and running AND get everything else done for the beginning of the fall semester. When things do not go perfectly, it makes the faculty and the students anxious (even if we fix it quickly). Goodwill, with respect to technology, is a precious commodity and must be protected! I will not recommend that we upgrade to 1.7 until June. We have many fewer people on the system then and more time to get after any problems that DO arise.
Why wait? It is less a version issue than a timing issue. Partly the timing, partly because we depend on some third party blocks and modules that may not be ready yet, and partly because 1.7 will surely require more instructor training (maybe less than I think, if all goes well!). In larger installations, you have more variables that can cause problems and the whole process moves slower. Moodle is on a very fast development cycle! And I love that! But this scale/implementation schedule issue probably contributes to the testing and debugging problem.
atw
I would think that release timing is part of the problem. Certainly with such a globally used application there will NEVER be a perfect time to release, but at least in the US we are very locked down to our school calendars (isn't everyone? the calendars are all just a little different). I would think that few IT departments in large installations would be very willing to upgrade in the middle of a semester, or even in the middle of an academic year. That means your testing pool is smaller.
From our perspective, 1.6 came out in August (wasn't it?). I went ahead with it because of the essay questions in the quizzes. That was important to us. But we use several 3rd-party blocks and modules. Some of which were not too hard to get into operational order (with a little help from our friends here on the forums!) and some of which we have given up on for now (upload and review, for example). We were struggling to get that up and running AND get everything else done for the beginning of the fall semester. When things do not go perfectly, it makes the faculty and the students anxious (even if we fix it quickly). Goodwill, with respect to technology, is a precious commodity and must be protected! I will not recommend that we upgrade to 1.7 until June. We have many fewer people on the system then and more time to get after any problems that DO arise.
Why wait? It is less a version issue than a timing issue. Partly the timing, partly because we depend on some third party blocks and modules that may not be ready yet, and partly because 1.7 will surely require more instructor training (maybe less than I think, if all goes well!). In larger installations, you have more variables that can cause problems and the whole process moves slower. Moodle is on a very fast development cycle! And I love that! But this scale/implementation schedule issue probably contributes to the testing and debugging problem.
atw
I can see timing as an issue.
We are a small high school with about 700 students and 10 staff using moodle. For me "timing" is a weekend when, if something goes badly, I have time to fix it as best as I can. I didn't hesitate at all when I found 1.7 was out since I really never experienced any problems with upgrades before.
I'm almost afraid to ask, but I wonder if this "large scale" testing in different configurations is way MS released Vista on such a large scale?! What better way to find bugs than to have lots of testers (back to Martin D's comment earlier).
I guess I shouldn't complain being one of the guinea pigs to test 1.7 on a "large" scale. After all I'm not always able to contribute to the project in programming ways.
Jeff
We are a small high school with about 700 students and 10 staff using moodle. For me "timing" is a weekend when, if something goes badly, I have time to fix it as best as I can. I didn't hesitate at all when I found 1.7 was out since I really never experienced any problems with upgrades before.
I'm almost afraid to ask, but I wonder if this "large scale" testing in different configurations is way MS released Vista on such a large scale?! What better way to find bugs than to have lots of testers (back to Martin D's comment earlier).
I guess I shouldn't complain being one of the guinea pigs to test 1.7 on a "large" scale. After all I'm not always able to contribute to the project in programming ways.
Jeff
In addition to the timing, I think that the upgrade had a little bit of a 'refactoring of the code', feel to it. Or rather that the ratio
Added functionality / Changes to the code
is, perhaps, subjectively IMHO, a little lower than in some releases. This is as much about the denominator as the numerator.
The numerator, "added functionality" is the cake that directly motivates moodlers to upgrade. This time the chocolate cake that I am aware of was: Ajax course content manipulation, roles, and per quiz feedback. I don't use roles, and alas Ajax did not quite make it into standard 1.7. So the major cake-factor was in the quiz feedback, for me.
However on the other hand, the denominator, Changes to code, were (I hear) quite substantial and this may mean a bit of a hassle. Changes to code mean that non-standard modules, addins and hacks are likely to have issues, and upgrade problems seem likely to be a little more prevalent.
Hence the bang-per-hassle-buck may have appeared to be a little lower.
Added to this, some of the added-functionality is likely to be atractive to new users. The fact that MSSQL is now supported is great for some new users, but for the existing (would be beta tester) users this is not likely to be an advantage. So however much improvement and flexibility went on behind the scenes, this seems likely to be less attractive than a nice new feature.
Every time there is an upgrade it helps if there is an emphasis on a banner, functionality-cake that users just can't afford to be without. This time I was not sure what the cake was. There needs, I feel, to be a drop dead cool new feature or two (e.g. messaging in 1.6) with each release to ensure that the moodler masses upgrade. Ajax was nearly that, but...Did it make it in there?
Of course other moodlers may be more altruistic than me. My attempt at 1.7 altruism was an installer including 1.7, to aid and abet testing but it fell dead from the press through no fault of 1.7.
I admit I am not into the front page admin menu either. But I would like to hear it from all 1.7 users on the big dollops of functionality that they are enjoying.
(Imagine Ballmer's "Developers" chant here) Roles! Roles! Roles! Roles!
Almost everyone I work with is positively salivating over roles. It's not a favourite of mine per se, but it means that we get to remove a lot of little old hacks we had to allow special things for specific sites.
If roles, DB support or Ajax don't don't list high for you, I can understand 1.7 is a bit of a "so what" release. If you didn't care for unicode, 1.6 was also not sooooooo hot. No harm in that either -- I've skipped 1.3 and 1.6 myself
Almost everyone I work with is positively salivating over roles. It's not a favourite of mine per se, but it means that we get to remove a lot of little old hacks we had to allow special things for specific sites.
If roles, DB support or Ajax don't don't list high for you, I can understand 1.7 is a bit of a "so what" release. If you didn't care for unicode, 1.6 was also not sooooooo hot. No harm in that either -- I've skipped 1.3 and 1.6 myself
Roles!
I lack imagination. I shall go and look at the Roles forum and try to get some ideas on how to use roles. Now I remember why I upgraded to 1.6 -- unicode.
Salivation is in the mouth of the beholder.
From what I have seen of the new admin panel, I hate it. Even though the old single page was long, at least you could find what you were looking for quickly by simply doing CTRL-F and then searching for whatever it was you needed. Now you have to open a bunch of little tabs, and they aren't necessarily intuitive.
I really would like to use roles, but I've had to work around the lack of them up to this point so it isn't something make or break at this point that I am willing to have emails being sent everywhere they aren't supposed to go.
I still haven't figured out the migration to Unicode properly on my site anyway. The first attempt was a disaster and I rolled back, and still haven't seen a good explanation anywhere why. I'm going to try again and investigate more carefully what went wrong and if I have to do so, I will simply manually convert the necessary fields one by one by one....
I really would like to use roles, but I've had to work around the lack of them up to this point so it isn't something make or break at this point that I am willing to have emails being sent everywhere they aren't supposed to go.
I still haven't figured out the migration to Unicode properly on my site anyway. The first attempt was a disaster and I rolled back, and still haven't seen a good explanation anywhere why. I'm going to try again and investigate more carefully what went wrong and if I have to do so, I will simply manually convert the necessary fields one by one by one....
What kind of problems did you get with the Unicode migration? I also had a lot of problems but don't really understand them clearly enough to explain what they were. I kept getting stuck in loops or ended up not being able to select the language packs correctly, or could select them but then had one language pack but was shown as having another, and various combinations of this (as in I use the language menu's, which would display Japanese, but show English, or visa-versa, or tell me a pack was already installed even though it wasn't, or install it but then not show it correctly, repeat ten times then fade . . . ) . And the characters weren't displaying properly. I was very kindly given a hack to eliminate this but unfortunately have not got far enough to even try it. The vast majority of sites seem to be monolingual, so I am wondering if these problems are restricted to genuinely bilingual sites (as in sites using more than one entry on the pull-down menu). I also have a serious hotch-potch of encodings on my site, which probably doesn't help. Am intending to use a partner to sort this out. Would try again myself but am still slightly traumatized from the last attempt . . .
(still a big fan of moodle though, it would have to actually jump off the server and shoot me for me to stop using it!).
(still a big fan of moodle though, it would have to actually jump off the server and shoot me for me to stop using it!).
Dont forget to use the admin search function. I was also a little upset with the interface until I learnt I could just type what I wanted to edit into the search field and have what I needed displayed.
Just type in RSS and you get all the rss options...very nice.
Would like to know exactly how the notifications space is meant to be used though
Just type in RSS and you get all the rss options...very nice.
Would like to know exactly how the notifications space is meant to be used though
Personally, I don't like the new admin setup from the main page.
Maybe I'm too used to the old way, but it doesn't "feel" as intuitive.
JM
Jeff-I had the same experience with 1.6, so I do not think it is limited to this release. I upgraded to 1.6 within a day of its release, and it was a very painful process. I was filing probably dozens of bug reports about it and patching left and right. But I could accept it at the time because I was still running my beta test with a group of students who knew to expect problems. However, I've now moved past that stage and have many full-paying students who expect things to work. So I cannot afford to upgrade to 1.7. I've noticed someone reported a bug with Paypal. If that were down it would cost me a lot because I wouldn't be able to enroll new students. I'm waiting for 1.7.1 or 1.7.2 even when I feel assured that others have found the nasty bugs and they have been fixed. Whether they are bugs, or simply the documentation needs to be improved, I think it is the nature of the beast that until Moodle is used in real life situations, these things will not appear. I thought some testing protocols were being developed and perhaps some focus needs to be put on that so these problems don't happen again.
Roles... have been interesting...
During upgrade we have had alsorts of roles assigned to all sorts of people... resulting in loads of e-mails being sent to many people whenever anyone submits an assignment... so I am going to strip them off and start again...
But I think that is a good thing... a long overdue tidy up does no harm.
I think roles will be brilliant once I have sorted it all out.
Antony,
This is almost my experience exactly.
Can I ask what you mean when you say "so I am going to strip them off and start again..."
I'm debating if I should remove all users from all roles and then reassign teacher roles on a per course basis.
Thoughts?
Jeff
This is almost my experience exactly.
Can I ask what you mean when you say "so I am going to strip them off and start again..."
I'm debating if I should remove all users from all roles and then reassign teacher roles on a per course basis.
Thoughts?
Jeff
I agree with ya.. Im not feeling 1.7 also
Dont get me wrong.. I feel if I had a choice between 1.6 and 1.7 , i would go with 1.7 all the way.. Im hoping that 1.8 has something nice to offer. . Either way. I love moodle all around and that goodness for all who put effort into it... Without it my site wouldnt be as great.. well it would be.. but it would be minus moodle..
Dont get me wrong.. I feel if I had a choice between 1.6 and 1.7 , i would go with 1.7 all the way.. Im hoping that 1.8 has something nice to offer. . Either way. I love moodle all around and that goodness for all who put effort into it... Without it my site wouldnt be as great.. well it would be.. but it would be minus moodle..
I second too. Here at SFSU we just did some interviews with our instructors who had intensive experience with both Blackboard and Moodle. Almost all of them pointed out that they like the simplicity and weekly format of Moodle (we are still running Moodle 1.5.x), and almost all of them mentioned that the current Gradebook is the weakest link. They also expressed their interest in Eluminate type of video/audio conference and a better chat tool (more scalable), which I know Alex Little and his team is working on right now (see: http://labspace.open.ac.uk/). IMHO Moodle should really pay more attention (and make it higher priority) to software QA, testing, usability, and make these tools better (gradebook, chat, virtual classroom type of tool..etc.), instead of rushing creating more and more features that users might not actually needs. But then this is just my humble opinion based on our university's needs...
I don't think there is a rush to create more and more features that no-one needs ... Roles is a structural thing that consolidates hundreds of different feature requests into one system - it'll make a LOT of things easier in future. Likewise the XMLdb feature simplifies things for developers and increases stability.
I couldn't agree more about focusing more on testing and QA, but sometimes we have to go where the funding is. Few people want to fund developers for testing and QA, so I need to find that money from elsewhere. The process for this has improved markedly in the past few months since we set up the tracker, by the way.
Currently in 1.8 our focus is totally on smaller usability issues, and a long list of fixes/cleanups (not new features) (see MDL-7396) because people are in fact paying us to work on that.
.
About the gradebook, the planned gradebook rewrite was last week costed at US$45,000. I really would like to see it done for 1.8 but we need to find that funding (or someone good to step up and volunteer their own time). Perhaps you can ask your university.
I couldn't agree more about focusing more on testing and QA, but sometimes we have to go where the funding is. Few people want to fund developers for testing and QA, so I need to find that money from elsewhere. The process for this has improved markedly in the past few months since we set up the tracker, by the way.
Currently in 1.8 our focus is totally on smaller usability issues, and a long list of fixes/cleanups (not new features) (see MDL-7396) because people are in fact paying us to work on that.
About the gradebook, the planned gradebook rewrite was last week costed at US$45,000. I really would like to see it done for 1.8 but we need to find that funding (or someone good to step up and volunteer their own time). Perhaps you can ask your university.
Re $45,000:
Wow... So much for my "if we would each chip in ten or twenty bucks" theory, huh?
I am sure that the price is fair, so that is not some sort of complaint or criticism. I am simply constantly amazed by my own naïveté when it comes to business.
Oh, the clueless, sheltered life of a simple high school teacher!
Regards,
Art
Wow... So much for my "if we would each chip in ten or twenty bucks" theory, huh?
I am sure that the price is fair, so that is not some sort of complaint or criticism. I am simply constantly amazed by my own naïveté when it comes to business.
Oh, the clueless, sheltered life of a simple high school teacher!
Regards,
Art
Just to be clear, that was not my estimate, it was a quote from a developer using fairly standard US hourly contract rates.
But yes, it can be surprising what Free software costs. Look at the project costs here: http://www.ohloh.com/projects/25
But yes, it can be surprising what Free software costs. Look at the project costs here: http://www.ohloh.com/projects/25
Hmmm... Let's see if I have a spare twenty million lying around.
-- Art
-- Art
Ohh.. how cool.
And looks like DFWikiteam is doing some job there...
By the way... when the wiki stuff will be stabilized and done, I can get (and manage) good and creative students to collaborate in moodle needs at a regular basis. Even to work in testing and Q&A.
Don't underestimate the power of dfwikiteam
And looks like DFWikiteam is doing some job there...
By the way... when the wiki stuff will be stabilized and done, I can get (and manage) good and creative students to collaborate in moodle needs at a regular basis. Even to work in testing and Q&A.
Don't underestimate the power of dfwikiteam
The high numbers for dfwikiteam might be because they have been adding each new version into contrib as a new project instead of updating their existing files ...
But I'm really looking forward to seeing the new wiki ready!
But I'm really looking forward to seeing the new wiki ready!
Hi Art, to put it in perspective, $45k is about what it cost for one institution to run Blackboard 'Enterprise' learning system for one year.
For that cost, thousands of Moodle sites around the world could have an enterprise level gradebook that would make it much easier to support different grading schemas used in different levels of education, features like activity locking, connect with other systems, more granular roles, modify the view based on personal/local preferences, etc.
For that cost, thousands of Moodle sites around the world could have an enterprise level gradebook that would make it much easier to support different grading schemas used in different levels of education, features like activity locking, connect with other systems, more granular roles, modify the view based on personal/local preferences, etc.
Hi, Michael,
Yes, I am sure that you are right about that. And the money would be well spent!
Here's the most amazing, wonderful thing to me, though: All or most of that will probably come anyway. The odds are that legions of selfless developers will share their knowledge and skills with simple Moodle-end-users like me just because they are generous, just because it is a kind thing to do.
For example, while I am sitting here typing this, I am using the Excel to GIFT converter that Timothy and others developed. It is so cool and saves me so much time that I would gladly have paid fot it, but I did not have to. It was gift. A really nice gift to anyone who could use it.
Of course, people have to eat. Even developers need money, and there is a limit to the time and genius that they can simply give away. I really, truly do understand that.
But just about everyone here seems to give what they can. Well, that has been my Moodle experience to date, at least.
I wonder what the monetary value of all the time and expertise that the community has invested in Moodle to date would be... A gazillion dollars?
Regards,
Art