Beta Hotfix 4 - Beta 2


Perhaps not the best thread for this, but it's the most recent OS topic so I'll use it. :) Is there any reason readme.txt files (via XFCE menu -> Documentation) open in Arora? It can take up to ten seconds to load one, it would be much quicker if they opened in Mousepad. Is the allowance there for future HTML readmes, perhaps?
 
Gruso said:
Perhaps not the best thread for this, but it's the most recent OS topic so I'll use it. :) Is there any reason readme.txt files (via XFCE menu -> Documentation) open in Arora? It can take up to ten seconds to load one, it would be much quicker if they opened in Mousepad. Is the allowance there for future HTML readmes, perhaps?
Yep, I'm sure I've seen it mentioned somewhere that it's for HTML support, so you can have clicky links and pictures in your READMEs. I don't often RTFM but I would also appreciate a bit of magic to check for plain text and launch mousepad if appropriate. Or maybe the system's file type associations be used somehow?
 
Last edited by a moderator:
SteveM said:
Gruso said:
Perhaps not the best thread for this, but it's the most recent OS topic so I'll use it. :) Is there any reason readme.txt files (via XFCE menu -> Documentation) open in Arora? It can take up to ten seconds to load one, it would be much quicker if they opened in Mousepad. Is the allowance there for future HTML readmes, perhaps?
Yep, I'm sure I've seen it mentioned somewhere that it's for HTML support, so you can have clicky links and pictures in your READMEs. I don't often RTFM but I would also appreciate a bit of magic to check for plain text and launch mousepad if appropriate. Or maybe the system's file type associations be used somehow?
Well, the PXML.xml file does allow you to specify what type of file are you referring on the info section using mime types. Of course, a separate matter would be if every pnd packager fills this tag correctly.
 
Last edited by a moderator:
Blue Ion said:
Well, the PXML.xml file does allow you to specify what type of file are you referring on the info section using mime types. Of course, a separate matter would be if every pnd packager fills this tag correctly.
Heh, just shows you how often I use the info tag ;-) If set to text/plain, does it get opened in mousepad? A bit of auto-detection if no type is specified would still be nice though, anyway.
 
Last edited by a moderator:
Blue Ion said:
I'll add mine too, now that I'm at it.
False alarm. Turned out I forgot to flash uImage as well, and the new sgx driver must have been disagreeing with the one in the boot image.
Now that I have flashed the uImage as well, everything works fine now.

xopher, check that you had flashed this part as well.

As for the battery, it must have gotten calibrated or something, as I woke up today, it showed 98%, what it must have been, weird.

SteveM said:
Blue Ion said:
Well, the PXML.xml file does allow you to specify what type of file are you referring on the info section using mime types. Of course, a separate matter would be if every pnd packager fills this tag correctly.
Heh, just shows you how often I use the info tag ;-) If set to text/plain, does it get opened in mousepad? A bit of auto-detection if no type is specified would still be nice though, anyway.
Nope, it always open in arora, which has the benefit it can handle pretty much every format you throw at it, though it is slow to start. "xdg-open" or just "open" could work here, but the former is not well configured, and the latter in simply not installed, so...
 
Last edited by a moderator:
I may or may not have found another PND acting up with this update. I wont be able to verify until I get home because I never ran the emu prior to this update and have read of others having issues. The PND is gngeo-w-gngeox. The symptom I am experiencing is what appears to be during the mounting process. There are no errors in the runpnd-gngeo logfile. What I do notice is what I do not notice...I created a folder for the BIOS files. Since the frontend stays running while you set the config I was able to look...in /mnt/utmp where the PND mounts with the other stuffs from appdata......the bios folder I created does not appear within the mount. Only folders generated by the PND are appearing in the /mnt/utmp folder and it can not see the bios folder I created nor can I search for it. When I can confirm if it works on my older SD OS extraction I'll confirm if the outcome is different most likely via an edit of this post or if there is some activity I'll make a new post confirming or denying it to be coincidental. If anyone else can confirm before I can please do.

EDIT: ooooh....lemme try the uimage and report back. :) thx blue ion...is that the one in the current flashkit? If not where to get?

EDIT AGAIN: figured out what to do to generate the bootf.tgz :)

FINAL EDIT...maybe: flashing uimage fixed my problem as well.....again thanks Blue Ion!! I'm pretty sure the gngeo thing is completely unrelated....disregard that I .....
 
EvilDragon said:
torpor said:
ED: are you going to include the new wifi drivers in this hotfix, perchance?

Depends on if notaz puts them in the GIT or not ;)
Isn't the point of a beta to make sure the things that are in the beta are working? If you keep adding more stuff, you'll have to release more betas to test and keep pushing out the actual release of the hot fix. While getting better wifi is definitely a good thing, shouldn't you lock down the hot fix sooner or later and stop letting new things in so you can find all the bugs with the existing hot fix and start preparing them for hot fix 5 instead?
 
Last edited by a moderator:
It isn't like new stuff is flowing steadily, it's coming in quick bursts, after that we get silence periods. So I think while things are flowing they should get in, and during silence final release can be made.
 
Except it wasn't released during the silence period. It went into beta until another burst of new stuff, and then another beta. Repeat 2 more times, now a third time. We've had 3 betas which I would have considered final hot fixes in their own right. We should be on HF7 by now, with HF8 adding the new wifi driver. Ok, maybe not that extreme, but from what I can tell, HF4 should have definitely been finished a week or so ago, instead of finding new things that can go into it. You've got to make the cut somewhere, do a real beta test, and say anything new goes into the next one, otherwise you'll be stuck with these bursts of changes and an always-beta HF4.
 
WizardStan said:
Except it wasn't released during the silence period. It went into beta until another burst of new stuff, and then another beta. Repeat 2 more times, now a third time. We've had 3 betas which I would have considered final hot fixes in their own right. We should be on HF7 by now, with HF8 adding the new wifi driver. Ok, maybe not that extreme, but from what I can tell, HF4 should have definitely been finished a week or so ago, instead of finding new things that can go into it. You've got to make the cut somewhere, do a real beta test, and say anything new goes into the next one, otherwise you'll be stuck with these bursts of changes and an always-beta HF4.
Hopefully this will be the last hotfix before proper repository support gets finally set up. The next hotfix should just be a script that adds the urls from the new repos.
Thanks to this, people should be able to subscribe to a stable repository with tested updates and fixes from a specific release, a testing repository with the latest new features and fixes, and perhaps a bleeding edge repository with the latest packages problematic packages like new kernels, drivers and the like straight from the angstrom repository.
Therefore a simple opkg upgrade, should do. Or just update a new component that you are interested.

Of course, hotfixes should still be released for people without internet connection (or with a flaky wifi) and for those that want to have everything in one, a bit like the service packs for windows.
 
Last edited by a moderator:
I for one will be sticking with hotfixes regardless of my connectivity status. If something breaks I want to be able to say 'well it worked on Hotfix x, but stopped working after hotfix y' rather than think 'omg how many updates have I installed since I last KNEw this worked?'
 
Back
Top