Hei Maarten, thanks for your educating mail. This is actually the first time i get an answer to why things are as they are, that i can understand. However, being as ridiculously optimistic as i am ;-), i don't give up that easily... ;-)
> > 1. Make bouncer universal and just maintain the offerings in the > > JavaScript. This is the non-ideal solution, but it would mean that the > > JavaScript only offers options that are indeed available. > > > What do you mean with making the bouncer 'universal'? I take it that bouncer is php? By 'universal' i meant covering all versions, languages and OSes that have available versions lurking around somewhere (of course, for that the NLPs would need to be instructed to put their stuff in standard places). So that no version of OOo exists, that isn't accessible from bouncer. This also includes that no link from a correctly tailored JavaScript to bouncer would end up dead. With that set up, you would have to manually tailor the JavaScript to represent the current release availability situation. I'm assuming here that bouncer could be made mostly automatic, so that it doesn't need to be maintained. Then the only maintenance issue would be the JavaScript which doesn't sound to bad to me, but also not very good. So lets look at the other options. > Nope. No PHP... No serverside scripting... that is what is creating a > end-user website so sluggish. Collabnet is a collaboration system, which > is running on openoffice.org . I see... :-/ > Within this system we are creating a > website that doesn't need the collaboration features Collabnet offers, > and would benefit from more advanced features that the Collab system is > seriously lacking (such as a good server side scripting mechanism... The > only place where some other systems are running is on the services > subdomain if I am not mistaken... So there go much of your ideas... :( Ok, i see that there are certain servlets. What language are they based on, in what environment are they running? > I know, but it you shouldn't have typed 'perfectly' ;) Hei, i wrote "well", not "perfectly". ;-) > >> I have promoted this type of button as well... but it resulted in a <ul> > >> inside of the button referring to several ways of obtaining OOo which is > >> imho definitely not the way to go. I would like it a lot to have such > >> button. And... > > ??? I don't understand. > > Except if that approach tried to cover too much in one button. > That was the point... P2P, CDRoms and download links were presented in > the same 'button' area. Sorry, but such an approach is bound to fail. Although, you make me thing... > I think one difference to consider is size. OpenOffice.org is >10 times > bigger in download size than Firefox, and the JRE makes that >11.5 > times. This is why CDroms and other methods (like p2p etc.) are somewhat > more important. You won't order a CD-Rom for 9MB, even if you're on > dial-up that's kind of do-able... (I think, I'm getting used to > broadband internet). Can't accept that point completely. If we'd be consequent about that, we wouldn't offer a prominent download at all, or at least that button on the front page would be a "order a CD" button rather than a download button. Yes, the size is a problem. One solution for that would be to offer an installer for download that fetches the necessary parts at installation time. But that's a hole different topic. I know we must count on that download because this is the only way we can really stay true to the claim that OOo is available completely for free. And that's why size can't be the point. > Furthermore Firefox is a browser, and that's it. > Everybody knows what a browser does, and browsers are considered to be > free, so nothing special. OpenOffice.org is an office suite. But for > free? Where is the catch? This, though, seems a pretty good point. Makes me think if a one-click-downlaod button is the right way at all. You're right...we don't offer a little one purpose tool that you need to download quickly when you need it. But on the other hand, we also don't offer take away boxes in shops and ordering CDs is just too unimmediate. So we are thrown back to the download option. However, this makes me think that guiding the user to a separate "get OpenOffice" page isn't such a bad option at all. So i'll focus on coming up with a better approach to that page. > If you have other ideas please post them... but I am afraid we already > doing much of what is possible. Maybe setting up another website besides > openoffice.org is a better idea... but I don't have the resources for that. Where is discussion about the main page happening, though? Is this the list for it? Really, the main page is bad. For one simple reason: it lacks user oriented resources (besides big attractiveness issues). > Problem is also that a lot of people > have opinions about how these things should be done which really slows > down decision making... (which also explains my tendency to create a > completely new website from scratch if you want to do things really well). > > Another solution, still ignoring point a, would be to create a > 'cleaning'-team, that seriously goes through all, and then really all of > the openoffice.org website. This small team would be able to set a > style, and directions, to which site owners should adhere too. > > Yes, a bit controversial since it is a bit anti-bazaar way of working... > but I think that is also how Mozilla got it done. My experience is that if you offer an attractive enough way of doing things, people will adopt it willingly. I.e., if a stylesheet would be created that is at the same time attractive, simple to use and comes with enough predefined stylings for the content of project pages, people will be happy to adopt it rather than doing the work of creating their own. So I'd rather propose a styling team who's task it is to create such a thing rather than trying to force styleguides on people which will only make you hunt after them. The idea is rather to convince that this styling team is doing good work. André.
signature.asc
Description: This is a digitally signed message part
