Why so many problems with database adapters?
I know this is a moot issue: "... since developing database adapters doesn't contribute to our revenue, we won't spend time on it." But currently the database adapters have problems. I have tried SybaseDAv2 and ZODBC. A few problems have been reported to the Collector with patches. But no actions taken on the part of Digicool, for a long time. This reminds me of Java's Bug Parade... which was rather frustrating... you reported a bug, crossed your fingers, and hopefully in 4 or 5 months it would get fixed... this after your commercial project already got killed. Yes, you fixed the bug, but I am no longer using Java. :) Is there another way of loosening the tight control of release of products? If Digicool does not have the man-power for taking care of release of some component, why not figure out a way for others to jump in and do the fixes? Why repeat the mistakes and shortcomings of the Java Bug Parade system? Is there a way somehow of linking in the product pages to some sort of user feedback? It will be more honest to let new users become aware of the problems of the current products and where they can find the patches. Currently, a new user happily download a product with the false hope that it will work properly. If there is some sort of user-comment (like Amazon's book review), then the user would know what problems there are with the current release of the product, and whether there are patches already made by someone. A publicly visible user feedback would also probably accelerate the fixes of products, be they made by Digicool or any outside people. I guess a message bulletin board linked to each individual product gives much better help than letting new users wade through tons of mailing list messages. (Product owners can remove nasty or unappropriate feedback messages, if necessary.) On one hand Zope advertises that it has database adapters to most databases, on the other hand we know that these adapters have problems that sit unfixed. Database is critical in most commercial applications, and frankly Digicool is risking a bit when mouthwords come out about the quality of the database adapters and the lack of support. regards, Hung Jung ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com
[Hung Jung Lu, on Thu, 06 Apr 2000] :: Is there another way of loosening the tight control of release :: of products? If Digicool does not have the man-power for taking :: care of release of some component, why not figure out a way :: for others to jump in and do the fixes? Why repeat the mistakes :: and shortcomings of the Java Bug Parade system? Is there a way :: somehow of linking in the product pages to some sort of user :: feedback? It would probably be somewhat helpful to DC if there was some way for the user community to participate in prioritizing bugs; or at least some way not to decouple the issue and the followups. There are at least three categories of "important" bugs. 1. Security fixes 2. Bugs affecting a large population of users in perhaps minor ways 3. Bugs affecting a small population in showstopper ways Number 3 is probably the kind you are talking about here. How about a ZopeBugs Wiki? Or some way for zope.org members to "vote," Slashdot-style, resulting in rating up the bug? Both are subject to programmatic abuse, of course, but even that eventuality would be a good indication of the level of someone's frustration. ;-)
you reported a bug, crossed your fingers, and hopefully in 4 or 5 months it would get fixed... this after your commercial project already got killed. Yes, you fixed the bug, but I am no longer using Java. :)
This is exactly what has happened to us with Zope, two of us here love it and are willing to work around/with the bugs, inmature interface etc. The rest of the team has voted to stop using Zope. So if Zope is to have a future, more attention needs to be paid to this issue. Remember once you lose the confidence / support of the users, be they developers or end users, no matter what you do after that it really doesn't matter. Our mistake here was using Zope version 1.x, several developers on the team acquired such a strong distaste for the product the very mention of Zope causes them to cringe. Just thought you should know... DR
"Daniel G. Rusch" wrote:
you reported a bug, crossed your fingers, and hopefully in 4 or 5 months it would get fixed... this after your commercial project already got killed. Yes, you fixed the bug, but I am no longer using Java. :)
This is exactly what has happened to us with Zope, two of us here love it and are willing to work around/with the bugs, inmature interface etc. The rest of the team has voted to stop using Zope.
So if Zope is to have a future, more attention needs to be paid to this issue. Remember once you lose the confidence / support of the users, be they developers or end users, no matter what you do after that it really doesn't matter. Our mistake here was using Zope version 1.x, several developers on the team acquired such a strong distaste for the product the very mention of Zope causes them to cringe.
It's certainly a manpower issue, we can't solve all the problems. Another issue however is communication. Could your developers get the knowledge of their dissatisfaction to DC? Keep in mind this is a two way issue, you have to work and we have to work for effective communication. This list is just out of control, and it's really your only conduit to us. I think Patrick is right about needing a Bug Wiki. We should try it, the only problem I can think of there being over crowding on the front page, because everyone thinks their problem is of the highest priority. Simple Rule, no bug reports on FrontPage, perhaps, Categories? -Michel
Just thought you should know...
DR
_______________________________________________ Zope maillist - Zope@zope.org http://lists.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://lists.zope.org/mailman/listinfo/zope-announce http://lists.zope.org/mailman/listinfo/zope-dev )
Hi Hung, Hung Jung Lu wrote:
I know this is a moot issue: "... since developing database adapters doesn't contribute to our revenue, we won't spend time on it."
But currently the database adapters have problems. I have tried SybaseDAv2 and ZODBC. A few problems have been reported to the Collector with patches. But no actions taken on the part of Digicool, for a long time.
Are you actually paying for support? If you are so depending on the solution, it would be the next logical step to take. If not, so we can all hope and wish, but not _demand_ to get the fixes :-)
This reminds me of Java's Bug Parade... which was rather frustrating... you reported a bug, crossed your fingers, and hopefully in 4 or 5 months it would get fixed... this after your commercial project already got killed. Yes, you fixed the bug, but I am no longer using Java. :)
Is there another way of loosening the tight control of release of products? If Digicool does not have the man-power for taking care of release of some component, why not figure out a way for others to jump in and do the fixes? Why repeat the mistakes and shortcomings of the Java Bug Parade system? Is there a way somehow of linking in the product pages to some sort of user feedback? It will be more honest to let new users become aware of the problems of the current products and where they can find the patches. Currently, a new user happily download a product with the false hope that it will work properly. If there is some sort of user-comment (like Amazon's book review), then the user would know what problems there are with the current release of the product, and whether there are patches already made by someone. A publicly visible user feedback would also probably accelerate the fixes of products, be they made by Digicool or any outside people. I guess a message bulletin board linked to each individual product gives much better help than letting new users wade through tons of mailing list messages. (Product owners can remove nasty or unappropriate feedback messages, if necessary.)
Nice idea. This could help a bit.
On one hand Zope advertises that it has database adapters to most databases, on the other hand we know that these adapters have problems that sit unfixed. Database is critical in most commercial applications, and frankly Digicool is risking a bit when mouthwords come out about the quality of the database adapters and the lack of support.
There are other companies which distribute software full of bugs and low on features and have very good success with it .-) Regards, Tino Wildenhain
Tino Wildenhain wrote:
Hi Hung,
Hung Jung Lu wrote:
But currently the database adapters have problems. I have tried SybaseDAv2 and ZODBC. A few problems have been reported to the Collector with patches. But no actions taken on the part of Digicool, for a long time.
Are you actually paying for support? If you are so depending on the solution, it would be the next logical step to take. If not, so we can all hope and wish, but not _demand_ to get the fixes :-)
This is a good point, currently on our radar we are putting time into the Oracle DA and a PCGI bug because of paid support. I'm having to suffer through an NT install today for just this reason.
Is there another way of loosening the tight control of release of products? If Digicool does not have the man-power for taking care of release of some component, why not figure out a way for others to jump in and do the fixes?
This is absolutly a key issue, but there is so much more to software development than just being able to check it in. This works great for those killer bugs your talking about though, if you have the patch, great, if you can check it in, better? How does the community trust the people who can check in? Do me have some sort of evaluation period? I can think of a dozen people off hand I would trust with write access, but a few of them I would watch. How can we watch? Is there workflow in CVS? Doubt it. Approaching the rat hole at every increasing speed. Right now people can send a patch to the collector, but it takes us time time test it and understand it, especially if it's a deep issue. Other people need to get ahold of the patch and test it too, we need better verification from people we trust then just 'works for me' if we can't invest the time to investigate each patch thorougly. The community really needs to do that investigation. They could probably do it better anyway.
are patches already made by someone. A publicly visible user feedback would also probably accelerate the fixes of products, be they made by Digicool or any outside people. I guess a message bulletin board linked to each individual product gives much better help than letting new users wade through tons of mailing list messages. (Product owners can remove nasty or unappropriate feedback messages, if necessary.)
Nice idea. This could help a bit.
I vote Wiki.
On one hand Zope advertises that it has database adapters to most databases, on the other hand we know that these adapters have problems that sit unfixed. Database is critical in most commercial applications, and frankly Digicool is risking a bit when mouthwords come out about the quality of the database adapters and the lack of support.
There are other companies which distribute software full of bugs and low on features and have very good success with it .-)
We want to stay pretty far away from them. -Michel
participants (5)
-
Daniel G. Rusch -
Hung Jung Lu -
Michel Pelletier -
Patrick Phalen -
Tino Wildenhain