DMOZ Redesign Project

So, here we go - version 1.0 of the DMOZ Redesign Project. The idea is to get suggestions from all you people out there on a new look for the rather tired and bland dmoz web pages. The main reason for doing this is that I personally can’t stand the current design and - who knows - the powers that be at Netscape may even adopt the finished product one day if enough people support it.

So get your ideas and suggestions in regarding layout, fonts, graphics, buttons etc., and I’ll do my best to incorporate them into the demo website. I think it’s safe to assume any new design will need to retain a green colour theme and probably feature some kind of lizard - they are kind of dmoz trademarks. I’m personally open to more radical change - but I’ll leave it up to you guys.

Version 1.0 is still very similar in design to the current format - thought I’d ease you in gently...

DMOZ Demo Webpage

:: Katapult ::

Flash plugin?

<outraged scream>

<trying to be gentle>Do you realize we have editors that use LYNX?

We need a page that works fine, just fine, for text-only browsers for color-blind or completely blind users on any ISO-standard-HTML browser, 3.1 (well, actually 2.0) or later.

We need that much worse than we need ANY POSSIBLE gratuitous graphics.

You forgot the comet cursor, gator plugin, and other spyware, as well ad banner spaces, pop-ups/unders, expanding contracting ads like yahoo/msn, and personalization for what content you want to see when you visit DMOZ.

Just because you can, does not mean you should.

DMOZ is fine as is [/thread]

katapult, why are you bothering? We're not looking for a redesign of the site, especially not one that (a) changes the logo and (b) adds extra crap like Flash. As hutcheson noted, we need (and want) the site to be as plain as possible. If you don't like the current design, then don't use the site. Feel free to grab the RDF and create your own custom designed copy of the ODP database with all of the bells and whistles that you'd like to see there.

A large number of editors are in areas where high speed access is not available or is too expensive. Any design that requires any amount of increased bandwidth is not likely to be welcomed.

Besides which, most editors would probably prefer that any avaiable technical support hours be directed to enhancements that improve and ease editing especially for those on dial up lines.

Remember, many users accessing DMOZ content are coming in through other routes the the dmoz website.

I wouldn't mind seeing lots of ideas for a redesign.

But, as others have said, whatever the added features, it needs to be deployable across a wide range of platforms with varying capabilities.

So, I'm certainly not against using Flash (etc) for some additional navigations or presentations, but all the material and paths to it must be available by other routes too.

Take a look at CC/PP as the future of tailoring webpages to users' needs and abilities, and think about a redesign that puts DMOZ in the forefront of that user-centric vision

Specifically, about the demo page, your HTML carries a large number of errors. It needs to state its version and validate 100% against that version. (That alone doesn't ensure full cross-platform deployability, but it is an excellent start).

Features I'd like to see:

• Breadcrumbs trail -- let me see were I've been (would probably require me to opt-in to accepting a cookie)
• "Volunteer to edit this category" -- replace this with something like "120 editors are responsible for this category [count the names in the path up the tree, plus editalls, and metas]. Volunteer to be its specialist editor"
  [/list:u]

I think it would be good to improve the 'public face' of dmoz.

I agree that dmoz is not really used by your average web surfer in the same was as they might use Google / AllTheWeb / Yahoo etc., but if a new updated design was produced to attract this market - would that be such a bad thing? There is room for improvement on the demo page, but it's a good start IMHO.

improve the 'public face' of dmoz ... to attract this market

Click to expand...

You forget that this is totaly not our market at the moment. And I doubt there is any interest from DMOZ to enter this market.

but if a new updated design was produced to attract this market - would that be such a bad thing?

Click to expand...

Possibly a bad thing. We just went through months of horribly slow editing because we were getting too much traffic on the single server we had. The upgrade took a while but be finally have several servers, including 3 public servers that act as proxies, pulling a single page from the main server only once every four days.

But this system could also be easily overcome by any significant traffic, and we are very unlikely to get any more servers for a long time, if ever.

Dmoz.org was never really meant to be a web surfer hub. We don't make any money on the site and never will, so each new regular user is really just a drain on our resources. It would be much better if surfers went to one of the data users.

I know, you're asking yourselves "what kind of site doesn't want to draw traffic to it?" But that's just the point -- we're not trying to draw traffic in to us directly. Utilitarian would best describe our needs and wants for the site and that very neatly describes the site we have. Why change it just for the sake of changing? Really, if you want to put a new face on the ODP, grab the RDF and make your own little slice of heaven with all the bells and whistles you could possibly want. More power to you.

Definitely a VERY bad thing!

1) Practically speaking, as theseeker notes, it would interfere with our mission.
2) Professionally speaking, it would be setting up to compete with our customers (licensees), costing us and them resources that we don't have, and that they don't want to waste.
3) And (I see this as the most important) giving such level of concern to the appearance of our factory/workshop room -- as if that were our only attraction, or our major attraction, or even what we thought would make us attractive -- is a fundamental betrayal of our ideals. We're about unique content -- the triumph of substance over style -- quick and efficient service -- shifting visitors away from our site as quickly as possible.

Almost five years ago, staff had a professional designer to polish up the original website. I'm afraid we were rather an unappreciative audience, artistically speaking: The simple reminder "a color-blind person can't use this" DID outweigh some cool color effects. The resulting blandness is deliberate -- if YOU want or need different colors for text, background, links, etc., on YOUR computer, OUR website will not interfere with your settings. And, as I mentioned, some of us DO use text-only browsing, and the site HAS to WORK that way. Really, if the website doesn't meet OUR needs, we don't care what it looks like. (And if it does, we still almost don't care.)

I know this is a very different attitude than your typical graphics-design-firm customer. But it is ours, and it is precious to us.

Amen!

Tough crowd in tonight.

I can think of several reasons why a redesign would be useful, so jumping hard on someone who suggests one is not a great way of encouraging the growth of a voluntary community whose own About page says:

"You Can Make a Difference
Like any community, you get what you give. The Open Directory provides the opportunity for everyone to contribute."

With comments like "outraged scream" and "Why are you bothering?", Katapult does not seem to be getting what s/he's giving.

A simple statement of why the current design is like it is would have been sufficient. Maybe that would have led Katapult into looking at some of the deficiences in the current design. Three obvious ones jump to my eye:

• Very many pages do not validate -- and you don't have to look far to find them -- Shopping is one of them
• Use of HTML tables (with no header cells, even) rather than CSS divs for layout is likely to make the site less usable to many people with old browsers or some assistive technologies
• Again, in terms of making the site universally accessible to people with biological or technological disabilities, I ought to be able to specify my preferences, like wanting to see the data in one, two, or three columns, rather than the default layout.
  [/list:u]
  
  They are not the immediate sorts of changes that Katapult was trialing on the first go. But who knows what would have resulted if the outraged screams had been a more polite "this is how you can give"?

I think that right when the initial discussion started in the other thread we made it pretty clear why we didn't think that it was a worthwhile effort to work on a project like this. He wasn't dissuaded at that point

I don't think that you can take the "You can make a difference" statement out of context like that. The next line is "Signing up is easy: choose a topic you know something about and join." We're not looking for people to redesign the web site, we're looking for editors.

Sunanda - thanks for the support, I'm gettin' beaten-up here!

I do of course take all the other points of view on board too.

It seems that the main concern is with any adverse effect that a new design may have on editing performance, speed and accessability. If text-only browsers, Lynx and so on are still widely used then the concern is justified, therefore the Editor sections such as the Dashboard should remain as is.

However, like it or not, dmoz does have a public face - and it 'aint pretty. I'd use dmoz for general surfing all the time if it didn't look so drab and I'm sure many others would too. Still can't see why this would be such a problem. If it's down to server performance issues, then someone have a word with the DBA's - get the public servers and admin/editor servers separated and ease the resource demand.

Can't help feeling that there is a big opportunity being missed here to raise the profile and popularity of dmoz in general. There seems to be an eagerness to dismiss the redesign suggestion outright before any possible benefits have been considered. Take Zeal.com for example, similar set-up to dmoz, but the design is contemporary and just looks like a more fun place to be. That can't help but encourage more editors to apply.

:: Katapult ::

If you are willing to spend the time to put together proposals, and listen to the feedback given, and build test sites, I don't think there are many who will stop you - it's (mostly) a free world.

At some point, the ODP staff may take notice of it. Or they may not.

The editors can give you no committment that the work you put in will make its way to the Public side of the ODP.

So as long as that is understood, those who have interest in this project please go ahead!

Beauty is in the eye of the beholder. Me, I personally don't like the look or layout of Zeal (and it certainly doesn't draw me in as a potential editor).

>> If you are willing to spend the time to put together proposals, and listen to the feedback given, and build test sites, I don't think there are many who will stop you - it's (mostly) a free world.

I wouldn't stop someone who wants to but I'd discourage it being done as a proposed change to the actual dmoz.org web site (and I'd discourage it being trialed here but that's just me). Please, feel free to take the RDF, put it on your own site, and build a brand new look for the data that suits your sense of aesthetics without worrying about how the actual dmoz.org site looks. You wouldn't try to redesign Google's simplistic design, would you? Oh, wait, maybe you would ...

>> At some point, the ODP staff may take notice of it. Or they may not. The editors can give you no committment that the work you put in will make its way to the Public side of the ODP.

Most changes to the look and function of the site are made based on requests from within. It's easy for someone on the outside to look in and think they know what would be best but sometimes you have to actually know the community and the community's needs and functions to really be able to offer acceptable suggestions. The suggestions that stand the most chance of being used in some form or another are the ones that actually offer an improvement in the functionality of the site for the editors (whether or not they also improve the functionality of the site for the users). Take, for example, thumbshots.org. Links to the thumbshots.org-enabled mirror of dmoz.org were added to the dmoz.org pages more because the thumbshots are useful to editors in helping us track down moved, changed, and expired sites than because they look nifty for the users. Had they not proven useful for editors, I doubt the links would have been added at all.

The suggestions that stand the most chance of being used in some form or another are...

Click to expand...

...the ones that involve no work on the part of our already limited resources.

If you can come up with a new design that incorporates all of the requirements that we have mentioned here (even if only one editor is using a text browser or WebTV browser it justifies a multi-platform design), and it can be deployed with no extra work for our very, very limited technical resource, then a) you are a miracle worker, and b) I would then say go for it.

If you use the site for any amount of time the way that the editors do, you come to appreciate the simplicity of the current design. It ain't broke so it don't need no fixin'

I don't think that anyone is trying to give you a hard time here or be negative about it, but this really does seem like a waste of your time and energy. Personally, I just don't see it happening anytime soon.

>> The suggestions that stand the most chance of being used in some form or another are...the ones that involve no work on the part of our already limited resources.

Yeah, that, too.

Kat,

Don't despair, a partial redesign can be "optional". There is another possibility to allow a user to have user choice loadable set of browser style sheets as "Skins", that aren't part of the dmoz core-look at all. This is very common and doesn't change the sites basic functionality, but gives some customization to the user.

Note this capability only works on certain browsers and the user must choose to apply the "skin". Anyway look up "skin and style sheets" to a couple of examples. (I'd name some I like offhand, but the traffic for the private sites could cost the owners a bit. I'd try W3C for some examples.)

About Style Sheets: CSS is a use the separation of the HTML content layer, and the style sheet as the final transformation of the presentation layer (or View). Style sheets "Cascade" in that multiple style sheets can be applied to the same page as "overlays" for each other.
Anyway there are tutorials on the subject, but you still have some dependence on the UC for implementing it across dmoz, (setting up a directory for having users CSS's as skins, and a setting to pull the accepted skins from; unless you wish to just change your view of dmoz).

A large font skin was applied to a couple of our sites as user viewing choices. (It boosts font sizes by 2).

Deep CSS magic is a bit out of my level of wizardry, so you may find better assistance elsewhere. I am a "less is more" visual GUI guy.


Best
JOE3656