"in fact most agree that a choice of sites is good"


I couldn't care less about having multiple front-end sites to fetch your software from. I.e. the more the better. It's the back-end integration that is important.


It would be great if Craig could update his store to also support the repository spec, though I really like the site made by milkshake. As a developer I wouldn't want to have to upload my software to multiple repositories. I can imagine that Craig wants to maintain control over uploads to his store. I don't know to what extend the repository spec unifies upload & synchronization at the back-end, but at the very least the alternative front-end sites should be able to pull the free (as in beer) software from the official repository.


From my own experience so far, the huge downside of the PND format is the difficulty of keeping your applications up-to-date (i.e. lots of manual work). I have considered switching to running my OS from SD card and abandoning the entire PND spec in favour of a regular repository system just to solve this.


The repository spec is not just nice to have, it is what we need.
 
Last edited by a moderator:
Apparently not.

Consider yourself in a very select minority here Blue Protoman

Not referring to myself; I know that it was designed with the Pandora in mind. But do certain others?
For goodness sake, I don't think this matters. They can all be made to work on the Pandora, and none of them are hard to use (although Craig's is the only one of the three that's not usable in Links2, though that would be very easy to fix). Craig's is designed for those who like to use their fingers, I think. I don't like to use my fingers, so there's no advantage for me, but there may be for others.

I couldn't care less about having multiple front-end sites to fetch your software from. I.e. the more the better. It's the back-end integration that is important.
Lots of us have been saying this. Lots of users, not just devs. (and anyway, if devs are unhappy, it causes a lot of problems for the users) I'm glad that some fine people like milkshake and Tempel are now working to solve this. And I think the best ideas will prevail in the end.
 
Last edited by a moderator:
Out of interest, does the repo spec have provisions for REMOVING apps?


And if so is it limited to just that site or the entire repo collective?
 
Out of interest, does the repo spec have provisions for REMOVING apps?


And if so is it limited to just that site or the entire repo collective?


could be a little more specific?


my website gives moderators and app owners the option to delete an application yes.
 
But can it report to the client that a app is deleted?
You mean like a remote killswitch for the application? I would strongly oppose that.


If during the update process it reports that a PND is installed which has been pulled from the repository along with the reason for removing it then I'm ok with it (as long as removal of the local copy is a decision made by the user).
 
Chocie is always good; let the 'market' decide.


Also note .. onyl dl.openhandhelds supports themes, pdfs, zips, etc; it must survive.


jeff
 
But can it report to the client that a app is deleted?

well once the file is no longer there the JSON data will be different as it will no longer contain the data for that game, that doesnt mean the game will be deleted across mutiple sources if other sites decide to use the repo specifications as well.
 
well once the file is no longer there the JSON data will be different as it will no longer contain the data for that game, that doesnt mean the game will be deleted across mutiple sources if other sites decide to use the repo specifications as well.
But it does sound like it could be used as an indicator in the event that a malicious PND is uploaded to one of the repositories and subsequently discovered and removed.
 
Right, so Midori being shit is my fault?
no, but this is. (compare to this). Yes, producing valids HTML documents actually help to work with browser light on features...
It looks to me that it's only complaining that one attribute value in one tag is missing surrounding quotes. Are you suggesting that fixing that one syntax error (that I would expect most browsers to cope with, as it happens) will make it look better in Midori?
 
It looks to me that it's only complaining that one attribute value in one tag is missing surrounding quotes. Are you suggesting that fixing that one syntax error (that I would expect most browsers to cope with, as it happens) will make it look better in Midori?
I havent tested (no acces to the source code...) so I cant tell.


What I was trying to say is that, if Craigx have put enough effort (shouldnt be too much) the app store would support it.


You knowuntil a few years all website supported IE6 as borked as it was because it was the default windows browser.


I guess a pandora specific website could at least support its default browser
 
I'll think you find it [http://apps.openpandora.org/ , remark] works fine with any browser which supports basic html standards, and if you try it on the Pandora most apps can be grabbed in two/three clicks without scrolling or using the stylus.
Sure, usability for up to 500 PNDs ought to be enough for everybody. Which is good because this way we don't have to care about the issues that would occur when having a lot more PNDs available on apps.openpandora.org .


In my humble opinion, if there's anything that apps.openpandora.org is not, then that would probably be "streamlined" (previously known as "simple with no space wasted"). The only halfway sane way to use it is to use the touchscreen (not everyone does that), and even that has its issues. For instance, put in a couple hundred more PNDs and have fun with never-ending pages (yeah, I know, there are categories - but the more categories, the more wasted screen space (with the current frontend), and more categories (of which there is only a limited number available) would only defer usability problems). Not only does apps.openpandora.org only really work on 800x480 touchscreen devices, it even doesn't really work very well there.


In fact, I don't see /anything/ special in apps.openpandora.org , which is probably the reason for why people want alternatives given the numerous minor deficiencies it has. So yes, people do indeed want several places. However, a prerequisite to this is that these places ALL have to be up to date. apps.openpandora.org could have been the "central" source to deliver up to date PND information, but unfortunately, it's not, which is a pity (and surprised me actually, since it's the option that would have given you most control in the long run).


What I'm trying to say is: basically, if you don't build in support for the PND repository specification, apps.openpandora.org will be obsolete in no time, and if you don't work on the frontend, other "stores" will soon be much more streamlined. In other words, if you don't redesign apps.openpandora.org , it will perish. As Esn said: "[...] the best ideas will prevail in the end."


"Christ on a broomstick."
 
Last edited by a moderator:
anyone else feeling this thread is kinda arguing over very little?


just let everyone get on with it. if it gets used then it gets used if not it will naturally die. survival of the most suitable answer to a problem, whether official or not.
 
anyone else feeling this thread is kinda arguing over very little?


just let everyone get on with it. if it gets used then it gets used if not it will naturally die. survival of the most suitable answer to a problem, whether official or not.
Seconded. Just let Darwins theory play out.
 
anyone else feeling this thread is kinda arguing over very little?


just let everyone get on with it. if it gets used then it gets used if not it will naturally die. survival of the most suitable answer to a problem, whether official or not.
Seconded. Just let Darwins theory play out.
This is true. The best ideas will prevail in the end.


If there is good planning and goodwill from the start, though, (and an attempt to understand and acknowledge the needs of others and the inadequacies in what exists) the process doesn't take as long and isn't as disruptive. And heaven knows, the Pandora has taken long enough as it is. :)
 
Last edited by a moderator:
All this nonsense about "being streamlined"... Really, it can't be more simple than Craig's version of the archive. You press "install" and automatically it downloads the file to your preferred location. Plus: large fonts, screenshots, ratings and reviews. Yah, same stuff other sites seem to offer... Anyways, you need to spend your time with the PND itself, not the website that offers it.


Only thing I would change with Craig's version is updating the frontpage a little more frequently. It still says: "29th May: Click to follow Pandora Apps on twitter."
 
yes but craigs website doesnt check to make sure the pnd that is being uploaded is valid, it doesnt make sure the required PXML data is there and the devs have no "real" controll over whats uploaded (someone gets paid to upload apps).


Not many Developers can be bothed to upload to the app-store because its not a pleasent expeirance to upload.


and yes it is all about the pnd which is why my site encourages devs to correct any mistakes.
 
Really, it can't be more simple than Craig's version of the archive.
True and that is exactly the problem. Staying up-to-date with Craig's archive is quite a hassle as you have to manually dig through it on an almost daily basis. The new site of milkshake has a back-end which can be accessed automatically from the Pandora over WiFi to update the PNDs for you.
 
I also like the design of Craigs site, but what I've been missing from the very beginning was PXML interpretation.


I also created some PNDs. All I need to do on the new site is upload them. No need to search for a screenshot, enter a description, a category, a title, a version number or anything else. It just grabs everything from the PND file, so as a dev, I don't have to do this twice.


It's very comfortable :)


I have to try the Updater on the Pandora, that also sounds pretty neat.
 
Back
Top