Installation System / Game Marketplace?


lulzfish said:
I would call it a "Program Manager"

*ahem*


50a.winbasics.gif
 
Last edited by a moderator:
WizardStan said:
Monk said:
WizardStan said:
So you're saying that people who find it challenging to click a link on a website will be able to click an icon in a standalone app?

Yes, I'm saying that people who find it challenging to browse to/find and navigate an ancient style web site and check the possibly hudnreds or thousands of entries for anything that they might need updating, it would be easier to have a custom app that does that for them in seconds for all their "installed" applications. A little bit more, but yes I am saying THAT.
*snip*
Another thing people are talking about is automatic updates, this is a feature I'd definitely like to see.
Which is one of the benefits I was alluding to (and someone else briefly mentioned, but was glossed over in the "make it look pretty and easier to install" posts) that a stand alone app has over web client. But you're the only person who has been discussing it as such. Nearly every other post has been about making new installs easier. I haven't missed the point, you're the only person who gets MY point, and the only person I wasn't directly arguing against, even if you didn't realize it. You just got caught in the crossfire of "web sites can be made pretty" and "it's the same number of clicks to install either way" arguments.

And I knew that, at one point. I think I get a bit too carried away when I'm typing too much and I TEND to assume that other people arguing for the same thing as me will not necessarily cover the same points simply because I have - so I fear that most others, like me in that post, will be floundering looking for what we think is required but hasn't been mentioned (or not mentioned in our favourite way).

And yes, it IS a useful technique to try and draw people out into thinking about what they are proposing (I do it myself) I think I just got lost in the sheer mass of to-and-from without obvious movement. Ah well! Badly fallable at times :(

WizardStan said:
HURRAY! I'm not totally crazy!


Well I don't think any of us were going so far as to say THAT.... ;)

calc84maniac said:
For people who say that the end users might not know how to use web browsers - how do you think they got their Pandora in the first place?
tongue.gif


It used to be said that, typically, 80% of users know 20% of the features of the software that they use. Especailly since OpenPandora hope to sell Pandora through 2retail" outlets (albiet probably Intenret-based shops rather than physical stores) people will browse online shops. That's not the same as managing installed programs by browsing a list of files in the archive. Also, I did actually (when money was not so tight) contemplate buying one for my son. I imagine other people may get them as gifts too.

Username said:
Quick question, will Mozilla Prism be available for the Pandora? If so, just set it to take you to the archive website and there. You have a stand alone application to view it from (automatic updates wouldn't work though).


Updates and removal/uninstallation are kinda core components of the whole "local app" concept.

atomicthumbs said:
lulzfish said:
I would call it a "Program Manager"

*ahem*


No - not that. Windoze has many misnamed things.
 
Last edited by a moderator:
Monk said:
No - not that. Windoze has many misnamed things.
Like the whole "windoze" os. I should be called windows in my opinion. Because it has... windows. :)

On topic, I think a web page suffices for now, but it should have ready APIs for more integrated solutions. A web browser, being today's multi-tool, lacks focus even if it can functionally do what a desktop application would. From a usability point of view, a web browser has too many options not directly related to what the user is doing, which in this case is installing applications. For example, when the user downloads a PND a broser asks where to put it and provides the entire filesystem as valid locations, whereas a desktop app would ask "save on sd1 or sd2?". This is just a simple example, but there are concerns like this. An integrated solution could also provide other services like mentioned updates, removal and such.

In my opinion, there should always be a web page to download applications from, but a desktop version should also be available for aforementioned reasons. Simple by default (easy desktop client for beginners), customizable by preference (different clients, web page), changeable by design (PND does not depend on such manager).
 
Last edited by a moderator:
There's a lot of differing opinion on how something like this might work already. What I was imagining was an application on the Pandora (not a replacement for the OS!) which is a kind of 'App Store' but really just more like an archive resource as it wouldn't be primarily focussed around making money.

The application would let you one click install software from the archive directly on the Pandora without the need for a separate laptop, browser or anything.

It would also do a good job of promoting applications and make it more easy for people to discover new software than trawling the internet and forums. (Perhaps top downloads, newest additions, Pandora team recommendations, stuff from specific coders, latest coding comp winners etc)

One of the main advantages would be that it would show you updates available for your installed applications, again to save a lot of web browsing and keeping up with specific news threads etc.

It could even introduce a user feedback and rating system a bit like the Apple App Store.

As a final (not primary) advantage, this application would be simple and clean with a good interface and not attempt to be an OS replacement, just a very efficient way to promote software, and let users browse for new stuff and updates.

It would not need to work via a browser, it would be an entirely standalone app. If there were desktop Mac & PC versions, they could act as a syncing and content management platform a bit like iTunes.

If developers sign up, they could take responsibility for adding their own stuff so it isn't a mammoth task for anyone else to maintain, just to keep it tidy and ticking over.

I'm not a coder, but I could help with graphics.

If the Pandora is the console this could be the box. (Toolbox, Gearbox, Boxer etc)

®
 
I would like to state that I'm not against the idea at all, I just didn't think it was necessary for the audience. Nothing wrong with a layer of sheen though.
 
Shirohagen said:
There's a lot of differing opinion on how something like this might work already. What I was imagining was an application on the Pandora (not a replacement for the OS!) which is a kind of 'App Store' but really just more like an archive resource as it wouldn't be primarily focussed around making money.


I think there's a lot more in common than there is different, and although things like being able to pay for apps are relative latecomers to the concept (I had not originally envisaged that at all - stupid of me, not looking ahead enough) I think the BEST thing would be for capable folk to get together and hammer out what it needs to achieve. Or even (god forbid!) a thread specifically designed to specify the scope of what's required in bullet form (something, in other words, that you can glance at and say "and add this" rather than pages and pages of long wordy lists).

Given the timeline, however, I suspect we're too late. I can't see how we'd get something agreed, built, approved and available on default installations within two months that would meet all (or the vast majority) of the goals :(

But perhaps that's not a bad thing? After all, as has been pointed out countless times, most if not all of the first batch will be going to people well used to doing things "the hard way" - if we can't get an "easy app store like" add/remove program manager installed by default/on the CD that ships with the Pandora until the second batch - would that really be THAT much of a problem?
 
Last edited by a moderator:
I keep hearing a TON of "why" but almost zero solid "why not"...

Now I consider myself to be pretty "open" in general and like to see all sides but this is reminding me of arguing with someone with deep rooted beliefs, religious or political maybe... there is always some kind of impossible to lose semantic battle or some "reason" that just can't possibly be explained because the opinion is so rooted in belief that change is never an option. Everyone is welcome to their opinion, but basically saying a suggestion is "wrong" is just strange to me.

I don't know, I just think there is a lot of hostility coming from nowhere here. I don't care what the "default" is, but in my opinion a dedicated application with software browsing and update checking is what I would prefer to use. Sure I could use the browser, but I would like to see a list of what is available along with what I have installed and what needs to be updated. Not just a list of things I can install. Which to me is what the browser would show. I mean, yeah, off-device that is pretty much the only option, but on-device I think this would be a very professional choice.
 
The heat seems to have been largely illusory, arising from an (understandably) agressive/reactive response to the debating technique that WizardStan was using in this instance (which was MEANT to draw out what the concrete benefits of such a system were).

I believe (though I have not checked in detail) that dflemstr has already put some effort into a potential solution that would go a long way to providing what most of us feel would be beneficial (indeed the only thing that I KNOW it lacks is the ability to include any commercial or donation aspects).

Vorporeal also seems wuite willing (and presumably able) to work towards a solution, provided that sufficient back-end support for the "better than a web site" solution is provided.

Personally, I'd love to get Vorporeal and dflemstr and ED (or Craig, or ED AND Craig) in a room to talk the idea through, because it sounds like a good one and IMHO the faster it could be put into operation, the better. Heck - a Pandora AND Windows AND Mac AND x86 Linux program (compilable for all, binaries provided) that could build a Pandora bootable SD with your choice of software could be fun to play with right now - get your SD cards ready so you can play the instant Pandora is unboxed!

Ummm... that last bit is a tad unrealistic though, I suspect :(
 
The last part is a tad unrealistic, although it's not unrealistic for one program to function on many platforms. What I'm envisioning would be built on Qt, so it would be easily portable to any platform. Simple architecture #ifdefs could be thrown in to allow for automatic discovery of SD cards on the Pandora, but on a desktop (or other non-Pandora system), would ask for the location at which the SD card is mounted. From there, the functionality would be, for all intents and purposes, identical. In theory, it shouldn't be a complex program, the complex part is designing the interface so that it's simple enough for the average user but has all the features needed by a power user. The web back-end would also need a significant amount of work done on it to provide a robust interface by which a package manager could obtain all needed information.
 
Vorporeal said:
The last part is a tad unrealistic, although it's not unrealistic for one program to function on many platforms. What I'm envisioning would be built on Qt, so it would be easily portable to any platform. Simple architecture #ifdefs could be thrown in to allow for automatic discovery of SD cards on the Pandora, but on a desktop (or other non-Pandora system), would ask for the location at which the SD card is mounted. From there, the functionality would be, for all intents and purposes, identical. In theory, it shouldn't be a complex program, the complex part is designing the interface so that it's simple enough for the average user but has all the features needed by a power user. The web back-end would also need a significant amount of work done on it to provide a robust interface by which a package manager could obtain all needed information.


Yeah - and the way Pandora progress has been going lately I fear that getting the back end in place and a front end (there needn't only be one in the long run) AND sufficient software up there (for testing purposes, you understand) AND have the results avaiable by the time Pandora ships... tho in time for batch 2 shouldn't be as much of a problem. I'm kinda curious though whether Craig et al would be interested, or if they are just married to the old File Archive "because it's there" kinda thing?
 
Last edited by a moderator:
Well, while this may be convenient, it still seems stupid. If people can't even go to a website and download a file to an sd card then this product is not for them.

But if someone does have a lot of time to kill and really wants this I suppose it wouldn't hurt.
 
Username said:
Well, while this may be convenient, it still seems stupid. If people can't even go to a website and download a file to an sd card then this product is not for them.

And, of course, that kind of lack of attention was why Stan was trying to get people to list what was wanted - because "going to a website and downloading to an sd card" isn't what it does. It's part of it, sure. A major part even, technically. Sortof. But that's rather like saying you think KDE and GNOME are stupid because anyone who can't use the command line has no business operating a Linux machine. Either elitist or simply missing the point :(
 
Last edited by a moderator:
If it actually does some proper package management (allows you to remove packages, automatically updates them, etc.), then it would be worth it.

Otherwise, it is literally not any more convenient than a website.
The KDE vs. command-line analogy is a gross exaggeration. The kind of exaggeration that takes people's eyes out. Be careful with those, man.
 
Username said:
Well, while this may be convenient, it still seems stupid. If people can't even go to a website and download a file to an sd card then this product is not for them.

But if someone does have a lot of time to kill and really wants this I suppose it wouldn't hurt.

You're obviously missing the point of this "product". It's not just a tool for installing PNDs. It's a tool for managing them. It allows for installation, removal, updating packages, and will hopefully provide a better interface by which users can find packages that might be of interest. While you could theoretically just look through the package list on the archive, a good search tool embedded within a local package manager could make finding new interesting packages to install an easier and quicker process. If the only purpose was to do things that could just as easily be done with existing tools, we wouldn't be having a discussion. We obviously see value in producing a local package manager. I'm sure you can think of some features a local package manager could implement that would give it an edge over using the web archive. Come up with some and let us know - this is a community project to improve on something that many currently feel is inadequate or lacking in certain areas.
 
Last edited by a moderator:
lulzfish said:
If it actually does some proper package management (allows you to remove packages, automatically updates them, etc.), then it would be worth it.

Otherwise, it is literally not any more convenient than a website.

Yup - the (a) file archive using a web browser could be designed to be a lot better/nicer/whatever without needing a specialist local program. I am not at ALL against making the File Archive "better" or producing alternative front ends to it - but by itself it just doesn't (can't) do all that I'd like.

lulzfish said:
The KDE vs. command-line analogy is a gross exaggeration. The kind of exaggeration that takes people's eyes out. Be careful with those, man.


Any lesser example (e.g. "to make managing programs easier) is met with derision along the lines of "two clicks in a local program are no better than two clicks in a web browser" - and miss the point. The analogy is extreme, but accurate - it's about makign stuff easier for more people. And it's not ONLY those who don't know what a keyboard is that benefit from a Package Manager - updates alone would be worth it IMHO. But as far as Internet-awareness goes, based on some lies (statistics) from a science magazine earlier this year, the following percentages of populations don't even HAVE net access:

USA nearly 30%
Brazil 94%
UK 31%
Germany 36%
China 81%
India 96%
Japan 26%

The VAST majority of these will never, could never, know about the Pandora unless is becomes a PHENOMINAL success. A few might be gifted a games machine as an xmas or birthday present I guess, but most would not. It's still some sobering figures to consider - the connected world is FAR from all there is, and that's people who have no net connection - it doesn't include people who have limited skills or confidence in the areana.

This community has a lot to answer for, including reminding me repeatedly of GOG.com. Wifey saw that site recently, when I was browsing my "owned", my "wishlist" and their catalogue, and she loved how _nice_ it looked. How "homely", in some senses - the "my account" area has the products I've bought in a graphically rich library setting - games on a bookshelf. The "sihlist" has a different but equally "rich" graphic setting, and their overall catalogue is different again. she wasn't doing serious shopping there, but just a quick look made her feel more "at home" than at other sites. The File Archive wouldn't get a look-in in terms of appeal, even though neither system deals with managing programs (just obtaining them). Yet she uses computers regualrly, and has net access both at work and at home.

That is, of course, only the gloss side which could obviously be improved on the File Archive - but it's an example of the whole ethos, the command-line vs KDE or text web-browser vs firefox or File Archive vs a Program Manager approach. The whole idea of a Program Manager (at least in this context) is to make life easier and nicer for everyone that wants it. Some prefer the older ways (I myself still drop out to the command line in Windows to do stuff surprisingly frequently) but "ease of use" isn't something to beat to death with a stick - computers were designed, from the beginning, to take the drudgery out of what we do.
 
Last edited by a moderator:
Okay, thanks everybody for the lots of replies. I've seen a lot of enthousiastic responses, and some less enthousiastics too. Everyone is entitled to an opinion ofcourse.

Some people think the Pandora will or should only be available for the enthousiasts, i think you should look beyond that. The Pandora is the ultimate mobile gaming console, and also the ultimate mobile multimedia and application device. There is no reason to think only enthousiasts would want to use it. Someone even proclaimed that the ordering process is so 'hard', it will not attract novice users. I think that the ultimate goal should be to have Pandora's being sold by the millions anywhere, including MediaMarkt, Walmart, Toys'r'Us and the likes. To achieve that, usability and simplicity are keywords that should apply to the Pandora. Therefore i suggested an extreme-simple programm for application installation/deinstallation and updating.

That should not mean that simplicity prevents the Pandora from being used in other ways. In fact, the 'open' platform of the Pandora is what i think that could make this machine a huge success, since anyone can develop for it. Having a huge file repository, coming from the 'open' scene, is just what could make the Pandora extremely successful. On a sidenote, i think the Apple ipod/iphone is the only real competitor for the Pandora, since it also uses a huge file repository (the apple store).

When i read throughout these (already) 8 pages, i think everyone agrees that the file archive should be more 'available'. Whether this could be done with a local application or by using a web-application (or changing the current archive for that sake), seems to have split parties. And yes, even when it's going to be a seperate application, the online repository would still be available, a new system should never be an absolute replacement.

What i also can conclude, is that everyone thinks an update manager would be a great feature. This could be implemented in the same application, or it could be a seperate one. I think implementing it in the same application would have certain advantages, but that's just my opinion. Ofcourse running an update application from a website, would be extremely hard since it needs to resort to local resources, so at least on that point a local application would be necessairy. I think we can all agree on that right?

Also, this kind of application should never intent to replace the OS, that should be silly. There is nothing wrong with the OS as far as i know. I never suggested that, i only suggested it should become part of the default installed applications, preferable part of the 'SIMPLE-OS' as is currently being designed for the Pandora.

I think we can also most certainly proclaim that whatever this application will be, it's on a too short notice to be on the first batch of Pandora deliveries, since these are scheduled for the end of Oktober 2009. Doing the complete development in less than 1,5 month, seems impossible to me. But as Monk points out correctly, that shouldn't be a problem, since the first batch is completely sold to linux/gaming/scene/hardware enthousiasts who will be able to use the system anyway. They could still install the application as it becomes available, they should know how :) .

I personally think the system of dflemstr would be a great place to start, since he already has the database and his system seems to be the backend to maintain that database and allow developers to upload their software. Especially the feature he describes that makes the screenshots by running the software on virtual machines, seems very interesting to me.


Since most agree something has to be done, i dare you to do it ;) . Are there any volunteers that would like to make such a project? If i recall correctly, i already saw some people posting that they would build it themselves, and i believe someone stated that he was already in the process of building such system?

For such a project we need coders and graphical designers. As i said in my OP, i think the design is what makes or breaks a system like this. Ofcourse we want to see all fancy features, like ratings, recommendations, change-logs, links to websites and support forums and stuff like that, but having all in the same overview brings back the same problem of the current file archive: a surplus in information and text. Browsing the archive should be easy, and all advanced features should be hidden under a button and only be shown when requested. At least, that is my personal view.

Are there any people willing to work on this project? Since i'm not a designer and my coding is limited to PHP, i think i'm of little use to the core development process, but i would like to include my services to discuss the features and the looks-and-feel of such a system, so as an consultant in fact.
 
Note that my project (the PND Manager) is put on hold for now; I'm in the middle of refactoring some components of it, but I could trim it down some more to make it even more usable by a Pandora-local installer app. No problem.

Also:
1. There are solutions out there already that provide installation front ends. Take a look at PackageKit for a great example
2. I could create a client application, no probs. I just won't have time, probably.
 
Jourdy288 said:
I understand what you mean Stan, but let's look at it from a different perspective.

2hxlhll.png

That perspective.
I still think that the iPhone/XBL style app store is worth it. I get what you mean about fixing up what we already have, I say we redo the whole archive so that it's easier on the eyes. Honestly, once that guy up there sees more than twelve characters, he's bound to freak out. What happens when he sees the archive!? I still think the app store is the way to go.

I hope the GUI that came to you was a bit more neat than this picture ;).
 
Last edited by a moderator:
I actually have no experience coding in Qt (yet), but I've used various widget-based GUI libraries before (and have written my own, in fact). I learn these things absurdly quickly, so it won't be an issue. I've looked through the Qt API, and it doesn't seem like an issue. I've got a bunch of ideas already for the interface design. I'm up for working on a local package manager. I'd rather not do this alone though - it would be nice to have some help. :D
 
Back
Top