Jump to content

Bookmark issue


Kiersolvd

Recommended Posts

I've tried to edit the listing of a cache that is in a bookmark, but the system returns an error. Anyone else experience this?

Yeah, just encountered it a few seconds ago, came here thinking it hadn't been reported and you beat me to it. Must be some problem that started tonight or it would have been reported long ago. Looks like all bookmark edits fail.

 

Edward

Link to comment

This issue is fixed but requires a fresh build to release. Currently we expect to release two weeks from today. I apologize sincerely for this delay.

 

Thanks for the update. I see now that a different error message--Server Error in '/' Application.--is being returned to an edit request now.

Link to comment

I've tried to edit the listing of a cache that is in a bookmark, but the system returns an error. Anyone else experience this?

I am having the same problem and I am ready for it to be resolved. Just finished a long working puzzle cache and need to update my bookmark entries so I can log the final cache. Wonder why it takes so long to fix....

Link to comment

Wonder why it takes so long to fix....

 

Because it requires a completely new build, which means releasing code now would also propagate the two dozen other fixes that are only half completed. The result would not be pretty.

 

The next release is scheduled for one week from today. Thanks for hanging in there.

Link to comment

Or to put it another way, if we were paying $300/year instead of $30/year, it could have been fixed immediately. Introducing a separate branch into development and deployment has serious costs. Most of us prefer the more cost-conscious approach.

 

Edward

Groundspeak releases hot fixes when they think a bug is seriously impacting the ability of users to find or log caches. Apparently editing bookmarks is not considered that critical.

Link to comment

Or to put it another way, if we were paying $300/year instead of $30/year, it could have been fixed immediately. Introducing a separate branch into development and deployment has serious costs. Most of us prefer the more cost-conscious approach.

 

Edward

Groundspeak releases hot fixes when they think a bug is seriously impacting the ability of users to find or log caches. Apparently editing bookmarks is not considered that critical.

 

There are some fixes that can be hotfixed and some that cannot. This is the latter. It has absolutely nothing to do with priority.

Link to comment
Groundspeak releases hot fixes when they think a bug is seriously impacting the ability of users to find or log caches. Apparently editing bookmarks is not considered that critical.

To restate Nate's explanation again, they have to consider not only the impact of the bug but the cost and risks of a fast fix. I'm not familiar with their exact methodology, but in software development in general, sometimes the problem can be simple, the fix be simple, and yet rushing in a fix outside normal procedures can be very expensive. It is certainly possible to set up development procedures which allow almost any fix to be rushed into production, but there's a very sharply increasing cost/benefit ratio as you try to incorporate more kinds of fixes into "out of line" fixes. When you have to create separate development branches in order to decrease the time-to-fix, the cost goes way up. It's also possible to increase reliability so that we almost never see bugs in production, but there also the cost rises very sharply as you try to approach perfection.

 

In addition, QA on out-of-line changes is very expensive. If only an isolated component needs to be changed, it's not unthinkable to take some shortcuts -- that's the hot fix. But if a full build is required, then a full release-level QA is also required. Software generally follows the principle that when you start pulling on strings, you discover that everything is connected to everything else.

 

Would I like to have this bug fixed today? Yes, of course.

 

Do I hear anyone saying they want the membership fee increased to $100/year (gold, not platinum) to pay for the kind of resources required for space-shuttle reliability levels? I haven't seen such a request, certainly not from me.

 

Edward

Link to comment

Thank you for that explanation, paleolith. You are dead on.

 

Since we're already off topic...

 

Some of you probably noticed we're on a ~4 week release cycle. That is 3 weeks of development, and one week of testing. We have a large team now and in order to keep things running smoothly we need structure and a plan.

 

Disrupting that plan causes ripples throughout the team and ultimately results in a net loss of productivity, even if in the short term we appear to be gaining. Sure, we could rush out a release untested and this one bug would be fixed, but experience has shown us that other bugs are just as likely to have been introduced elsewhere.

 

The only time it is safe to break our release cycle is if instead we delay the release. In this case we can create a branch and the devs will be able to safely continue into the next development cycle.

 

I refer to this last point because I must confess that I misspoke earlier about releasing next week. We will be in the middle of an office move then(!). That kind of chaos is exactly what we don't need so the release will be the following Tuesday. I can't say how sorry I am that I misled you!

Link to comment

Thank you OpinioNate. Yes, I can relate, software management is a tricky and difficult business. And a release in the middle of a office move? Only a mad man would do that. Thanks for the feedback you have been giving and hope your new digs are nice. Perhaps I'll have to schedule a trip to see them.

Link to comment

I don't need paleolith to lecture me on software developement. I've been doing this for 30+ years on several large projects. I will apologize to OpinioNate (and the Grounspeak development team) for implying that this bug isn't being fixed because it is seen as low priority. He had indicated several times that the nature of this fix requires it be done as part of a full release instead of as a hot fix. That may even indicate that a hot fix was considered and ruled out.

 

On projects I've worked on and being involve in the change control process, defect reports usually have a severity assigned by the person opening the report. It indicates the impact that not fixing the bug has on this person's ability to get their job done using the system. The defect report is then assigned to an analyst who will look for the cause of the problem and make an estimate on what it would take to fix the problem. The analyst or a change control board will assign a priority for fixing the problem. Those problems given the highest priority will be assigned to a developer to fix and then to a tester to test. Finally the changes are incorporated into a new release. Priority is review from time to time so that low priority fixes can be bumped up so they don't remain bugs forever. From time to time, it has always been the case that management will decide that some bug fixes are so important that they require a hot fix or a patch to be delivered out of cycle. Paleolith is right that that this causes some problems in that a new branch be created against the latest release version to do this work. After the patch is tested and released, those changes have to be merged into the code that is being produced as part of the normal development process. We have tools that compare the changes in the hot fix/patch files to the mainline development files and often this merge can be done automatically, but sometimes a developer has to make additional changes so the fix will work with whatever other changes were being made. In any case the fix has be tested a second time to make sure that it still works. While there is some cost and while developers don't like having to drop what they are working on to make a hot fix, I don't think it is as bad a paleolith makes it out to be. My original comment was that we often see hot fixes done by Groundspeak. It is just that this isn't one of them.

Link to comment

I have never done any computer programming. I am like a native in a pacific island cargo cult, and Groundspeak is the airplane dropping cool stuff on me out of the sky. If they tell me it will be awhile before the next display of magic, I will say, "Great! Thanks a lot!" And no longwinded explanation of how airplanes and parachutes work and how they could have rerouted something to get me my stuff sooner would impress me very much.

Nate, I think I speak for unsophisticated simpletons everywhere :laughing: when I say, "Thanks for all your hard work, you make a cool game even more funner!"

Link to comment

I have never done any computer programming. I am like a native in a pacific island cargo cult, and Groundspeak is the airplane dropping cool stuff on me out of the sky. If they tell me it will be awhile before the next display of magic, I will say, "Great! Thanks a lot!" And no longwinded explanation of how airplanes and parachutes work and how they could have rerouted something to get me my stuff sooner would impress me very much.

Nate, I think I speak for unsophisticated simpletons everywhere :) when I say, "Thanks for all your hard work, you make a cool game even more funner!"

:anicute:;):(

 

Me too. This temporary downtime has made me realize how much I do appreciate and rely on the bookmarks as "works in progress" for planning upcoming road trips, puzzles in progress, challenges in progress... Certainly there are workarounds but I'll be happy when the care package includes this fix!

 

Thanks.

 

Bean

Link to comment

I've tried to edit the listing of a cache that is in a bookmark, but the system returns an error. Anyone else experience this?

 

I'm having the same issue, the only way to change a bookmark description is to remove the cache from the list and re-add it with the changed text description.

 

According to the issues quidelines:

OS: Win XP Pro (running on a actual machine)

Browser: IE 8.0.6001

Link: My bookmark list: http://www.geocaching.com/bookmarks/view.a...d5-6c6b6849e8ca

Link to comment

Yes, I'm still having problems too. Everytime I try to edit, it gives me an error message and knocks me out of the geocaching website altogether.

 

Unless something has changed this should be fixed when the next code release goes live. AFAIK that has not happened yet and I believe Opinionate's last post in this thread indicates they're hoping to release tomorrow.

 

"I refer to this last point because I must confess that I misspoke earlier about releasing next week. We will be in the middle of an office move then(!). That kind of chaos is exactly what we don't need so the release will be the following Tuesday. I can't say how sorry I am that I misled you!"

 

Posted on 10/21, so I think that would make tomorrow the aforementioned following Tuesday.

Link to comment

I have never done any computer programming. I am like a native in a pacific island cargo cult, and Groundspeak is the airplane dropping cool stuff on me out of the sky. If they tell me it will be awhile before the next display of magic, I will say, "Great! Thanks a lot!" And no longwinded explanation of how airplanes and parachutes work and how they could have rerouted something to get me my stuff sooner would impress me very much.

Nate, I think I speak for unsophisticated simpletons everywhere :rolleyes: when I say, "Thanks for all your hard work, you make a cool game even more funner!"

 

I'd like to nominate this post for "Post of the Year"

 

Freaking brilliant! I love it! Reminds me of "The Gods Must Be Crazy"

Link to comment
Guest
This topic is now closed to further replies.
×
×
  • Create New...