Suck on SkinsTuesday April 11th, 2000TheUIGuy wrote in about a Suck.com article that talks about Mozilla/Netscape 6 and it's skinnability. It takes a very negative view on what we feel is a very positive technology. Skins allow the user to pick their interface. Packages allow them to extend it. This customization allows anyone to choose how they browse the Web, manage mail, or use any of Mozilla's other applications. The author of the Suck piece, Greg Knauss, seems to feel that choice is bad, because there is the possibility for poorly designed or useless skins. That comes close to saying, "there could be ugly pages on the Web, so shut it down." Greg also states, "by adding in all the flexibility of XUL, the Mozilla programmers have removed our ability to make the application use the native controls of the operating system." This shows a fundamental misunderstanding of Open Source. The direction Mozilla is taking doesn't prevent anyone from doing anything. The capability to do exactly what Greg wants is right there, in the code sitting on the CVS server, and in the mind and will of someone who needs something done differently. Mozilla might not be doing what you want, but that does not mean that you are prevented from doing it yourself. (FYI, there are projects for Windows and Linux -- and nothing preventing a Mac project -- for embedding the HTML renderer into a native wrapper application.) This never ceases to aggrivate me. They take a very poor understanding of a concept, somehow manage to completely prevent any thought about the subject (what the hell does he think the word EXTENSIBLE means, anyway!?), and then go off and yap and spread their misinformation to more uninformed masses, transforming then into more MISinformed morons... And Maximum Linux almost did it with their new issue. They talked about the lack of browsers for Linux, trashed the Netscape browser, and made no mention of Mozilla until the third page, which is squeezed under another article, and easy to miss. I think Mozilla is really getting a raw deal. the big sites (like ZDNet and such) just gloss over thewhole project... I do think that Suck.com is right in criticizing Netscape for their new browser's appearance and UI design shortcomings. As developers, we all know the virtues of Seamonkey, so they don't need to be repeated here. However, I disagree with Chris on the notion that Open Source software like Mozilla is immune to end-user criticism just because anyone can enhance it if need be. Cars can be tweaked, too, but that doesn't change the evaluation criteria. Netscape has just released the first preview release of their next generation browser. Netscape hopes Communicator 6 will replace earlier versions on millions and millions of desktops, soon, so this is the final hour to voice opinions about the direction they're taking. I, for one, hope that Netscape will direct significant amount of resources into making Navigator 6 the easiest and most efficient to use browser out there. If that is to happen, then pr1 is farther from the final than a Netscape preview release has ever been mka wrote : << I disagree with Chris on the notion that Open Source software like Mozilla is immune to end-user criticism just because anyone can enhance it if need be.>> did Chris say that ? No. Mozilla is not immune to criticism, we all know that, but before people criticize something, they should fully understand what they talk about. Chris is just explaining what Mozilla really is... and it's hard because it is a rich project with lots of new and strong ideas... It seems that the mozilla.org project has so far served the developer community and denied what the common users need in a modern BROWSER. What USER needs something that was supposed to be a browser but is now primarily called an OS by the people who created it? I would have served the USER first and then the DEVELOPER, but mozilla.org did it all backwards. Why? While criticism of Mozilla is certainly okay, and even useful, the impression I got from the article was that Chris had an axe to grind with Object-Oriented Programming, and wasn't inclined to give an objective evaluation of an OO project. hmm... the techies always defend skins by saying "well you can change it to suit your desires" but this is inane. the problem with skins is that it has burned countless brain cells and is arguably, completely useless. it's ridiculous how a menaingless feature such as skins can dominate the discussion pushing things like getting something launched in a reasonable amount of time to the back burner. again, the problem with the mozilla project is the horrendous scope. it should not take 3 years to build the thing that everyone wants which is a high performing, standards-compliant, lean, stable rendering engine/browser. lose the skins. lose the editor, mail and news as well. they have no business burning up the super-limited developer resources. It's only been TWO years since Netscape released the source for what was then Communicator 5. The current Mozilla has only been in development since October (or so) of 1998, about 19 months. Long for a software project, but it's open source, with hundreds of developers from around the world. As they said in 'Code Rush, it's like herding cats. It's been a long ride, but they've covered some good ground. Of course, I think October is a bit longer than a lot of people would like to see. When M14 hit, I figured M16 would wind up being the Release. I'm starting to doubt that... >> they have no business burning up the >> super-limited developer resources. Netscape's or other's? If people want to work on skins support for skins or whatever, who are you to say they shouldn't, or it's a waste of their time? They don't work for you. Take the code and make it like you want it; stick on a native lean ui for your favorite OS. If you're not willing to do it yourself, you have very little to complain about. please read www.mozilla.org/status/ (do it really, please) and you'll see that skins does not suck development resources... skins are only a portion of the development effort. but maybe the most over-looked portion, though. When I use Mozila to surf the web I really dont notice the chrome. I almost never notice the chrome on any app (unless if the app is 80% chrome). Thats not where the work gets done. While I know that the word "chrome" in this context has a historical background in software much less customizable than Mozilla, I find it nevertheless fitting when a developer calls the user interface "chrome". It goes a long way in explaining why the current "chrome" of Mozilla is sub-par in every imaginable way. That's why the interface needs to behave like all the other apps on the OS. You *shouldn't* notice it. But Mozilla can't be ignored. It just doesn't have appearence or behavior consistent with native OS apps. We can debate the merits of that until doomsday but it doesn't change the fact that "skins" or Mozilla's interface is different enough from other OS apps to deter adoption by many mainstream computer users. This just goes to show the power of the Internet: any idiot can write any crap he wants to and have it read. ...was pretty much on. I feel that as a user, I should at least get the option of a UI which is consistant with the native GUI. I like things to be consistant. I really don't need whiz-bang themes/chromes/skins, I want to be able to sit down use the app without having to think, "Okay, now is this a button or a menu? What happens if I click here?" And, quite frankly, it's more work for me to make things all look the same. Will I have to get a GTK skin for mozilla, or a mozilla skin for GTK? When you say, "Skins allow the user to pick their interface. Packages allow them to extend it," You're completely ignoring that the average user doesn't really care how nifty XUL is, nor are they going to roll up their sleeves and extend it themselves. They have better things to do and they trust that the people who write the program that they use (in this case Netscape/Mozilla) will look and feel like what they're used to in other applications on that particular platform. "[...] Greg Knauss, seems to feel that choice is bad, because there is the possibility for poorly designed or useless skins. That comes close to saying, 'there could be ugly pages on the Web, so shut it down.'" No, to me he seems to be saying (rather long windedly) that "Sometimes more is just *more*" You see I really like Mozilla, and I think that themese are great, but they should be left down at the level of the toolkit. My half-drunk $0.02, David Ironically, your interpretation of the article is very much different from mine. My interpretation is that he meant that users should have no choice whatsoever; that everything should look like everything else, regardless of what the user wants. An option to look like Netscape 4.x is good. No option to look like anything other than the underlying operating system is bad. As the average user who doesn't know a thing about C / XPcom / etc, i think that first Mozilla should have worked on a skinless Mozilla and fix the browser & mail bugs - of course, at least having a MINIMUM base for a decent, Netscape 4.x-compatible user interface (i did read about the platform/UI compatibility problems). While so much time is invested on the skins, many users (i suppose - if i´m wrong then correct me) are desperate for the 100% PNG support (bugs 16742,3013,3195,19283,28616), or a better low-memory support for earlier 16MB 486's (mine has 16Megs). On the other hand, if developers come in and they WANT to work on the skin stuff, who am i to tell them NOT to work? It's a free resource and it should be appreciated. Why not beat IE on this? At least MS can't say - hey, we gave skins to the webbrowsing stuff, you owe us :) BUT - and here comes the important part: If developers work both on skins AND Layout engine / MathML / PNG etc, then priority should be given to the latter. What we (at least me) want is a bug-free, standards-compliant browser with optional features (i.e. skins) available for a later version, not a bells&whistles browser still in beta-status because there are many bugs to be fixed - i mean, that's precisely why i think Moz became open-source. Building a native "classic" skin would take alot more time than building it using XUL/XBL/javascript. I've made attempts to build one using XUL about 2 months back, but I gave up because Moz was not that skinnable yet. If I find more time I'll try to clone the 4.x UI again soon. That is if no one beat me to it. ;-)
As a regular reader of suck.com I came away with the impression that most of the mo'zine readers treat suck as a place for tech commentary. Really, they're only job is to make fun of everyone. Especially Canadians. So don't worry, Mozilla is in good company as far as suck is concerned. You should see the vanity of skinning (and I guess it is sort of vain) as a part of the larger vanity of boredom-from-surplus that suck satirize for a living. No, it's not. Suck simply rips into people that need ripping into -- at least, in their eyes. (And if anyone was reading the user reviews of NS 6, they'd understand this.) But the point is still true. Skins destroy the basic point of user interface, that all things should operate equally and are easily identifiable. Since XUL allows you to change pretty much anything, it means that mozilla won't act like any other program you have. This is a reason why there are interface levels, like the Appearance Manager in MacOS, so that all programs share the same look so that an overall familiarity with a program is established. And might I note this is ESPECIALLY important for Mac users, as the functionality of the Mac UI is what attracted them to the platform... and dare I say there are just as many fanatic about it as well? An enormous amount of user reports turned out a near zero rating for the app, saying it operated like "a Windows/Linux program". Frankly, Netscape screwed up big time by releasing a PR build WAY too early. By the time the "native OS" chromes are out (which BTW should be the default install on a computer for the sake of consistent UI -- so far a losing battle on the Net) and the builds are fast enough to not make the skin layer noticable, hopefully people won't have already set their mind against 6. the irony is that the outside chrome such as aphrodite and sullivan look 10 times better than Netscape 6's and also looks much better than 4.x to boot. I still hope Netscape will go back to their senses and adopt one of these UI designs instead of trying to convince millions of people that off-centered grey is beautiful I think one of the problem with the misunderstanding of XUL is that it is thought of as chrome and skin. In truth it is more than that. Rather than calling it chrome and skin, I'd prefer to call it an application (or package) with customizable interface. Okay, back to more practical stuff. I heard that there were efforts to get CSS2 system colours working on certain platforms. Haven't heard anything about that in a while, anyone know what is going on on that end? What about the CSS3 draft stuff? Not standard I know but XUL really needs the "UI for CSS3" section 4.3 (fonts) and 5.2 (Key equivalents) FONTS http://www.w3.org/TR/2000/WD-css3-userint-20000216.html#font KEY EQUIV. http://www.w3.org/TR/2000/WD-css3-userint-20000216.html#key-equivalent So, what is this project that is embedding the Mozilla renderer into a native Linux (I presume GTK+) application? Will it be scrapping the Editor/Mail/News/Other shite I don't use? Mozilla is not just "one" project. There are many projects under the Mozilla umbrella. Linux haven't taken over the world yet. Even if it does, a version of Moz that uses XUL is still needed. Not that I don't want a version in GTK, but XUL has too much potential to be dropped. If you really want to see GTKMozilla become a viable alternative, it's time to lend a hand I guess. Here is the link if you need it. http://www.lysator.liu.se/~alla/gtkmozilla.html I am inclined to agree with some of these people who have their heads on strait as far the browser development is concerned in that I don't care if the entire UI of Mozilla/Netscape is hot pink so long as it will work without crashing every five minutes and will speed my web browsing. Yes, at present the Mozilla/Netscape effort has a ways to go as for as bug fixes but what I have seen so far is exceptional. No, it still isn't quite as stable or bug-free as earlier versions of Netscape, or even IE for that matter, but it is on the right track. I am so looking forward to the day when I can run my Windows based system without IE embeded in the system to screw things up and seems that Mozilla/Netscape will do just that. There is (and will continue to be, until we come up with a new term) confusion as to what is a user. In this thread, the term is used to mean two different things. First, the person doing the browsing. In this case, I am of the opinion that skins are utterly useless. The average person will not want to take the time and effort to change the skin. In addition, the average person would probably not be able to do it anyway. Second, the person designing a user interface (AlphaNumerica for example). In this instance, skins can be invaluable. The sullivan skin is by far the best looking browser I've yet to see. As a developer currently working on the architecture for a web based client for our application, I am paying very close attention to these skins. Even though the powers that be wouldn't even consider such a project, when I show them our web pages in a browser that makes the whole window look like our app, I think they'll jump at it. In other words, it won't look like you're just browsing. It will look like a real application that just happens to have some of the same behavior as a web page. Both of the groups are users. We just need to be aware that some users are more equal than others. Have a nice day! Mozilla doesn't have skins as an optional feature. Skins are not some side project that is stealing time from other areas of development. Skins are an end user bonus byproduct that result from Mozilla's basic design. The fact that Mozilla lays itself out rather than relying on the OS makes the widgets easilly styled with css and images. Every app under the sun has a skin, Mozilla just makes it easy to alter that skin in a simple and standards compliant way using css and images. There is no siphoning off of core development hours that yields this functionality. This is functionality that comes from a well thought out approach to building a cross platform application with a consistent look and feel. (and don't forget the i18n and localization benefits, the english speaking world is not the only world) Asa is it to add a feature to skin design to use a native widgets? If it is not difficult (i.e. takes less than a 2-3 man months), then by all means, it should be done. For the sake of speed at least. Even for the sake of not having to deal with the negativity of articles such as the one at suck.com I'm not too sure whether it would speed things up or slow things down, since Mozilla was not designed to work with them. I believe the performance issue has nothing to do with Mozilla using XP widgets or native widgets with the skin. That I believe have more to do with optimization.
I read the suck article and then had a beer then read it again. Somewhere in his rambling attempt at net poetry the guy has an axe to grind which is his right. I have my own misgivings about XP widgets relating to speed. However, the point was not missed by this guy but merely avoided. The reality is if you want your browser looking like every other grey app on your desktop then you will use the OS lookalike skin which will be everywhere. Checkout the MAC OS Skin in development in the SkinZone. The point is that Mozilla is visionary in its view on the use of standards, on compliancy and on the need for cross - platform consistency. Whilst XUL may have initially been a means to save CPP coding time it has has led to a whole new ball game. Greg Knauss sees the big picture for sure but chose to take the lazy approach and latch onto a line of attack devoid of any research into the product. Wonder does he "defend the right to innovate"? ;-) The suck.com article wasn't misguided. I like XUL, largely because all previous versions of Netscape appear to have had the UI done by a demented crack-monkey. Still, the default skin shipped with Mozilla looks like an afterthought and it's probably one of the better skinned-apps I've seen - most skinned apps ship with defaults which are bad enough to warrant a permanent uninstall just from an aesthetic standpoint, let alone their crimes against usability. What bothers me is that Mozilla is such a nice piece of code packed in an interface which is most charitably described as uninspired. It's like producing a sports-car that does 0-60 in 2 seconds while getting 200mpg and then making it look like a Yugo! Mozilla should use the native widgets by default for three reasons: - Consistency - users expect controls to look AND BEHAVE the same as they do elsewhere on the system. This includes consistency with a user-selected colorscheme and any system-configurable behaviours. - Compatibility - I *HATE* Mozilla's lack of decent keyboard navigation, at least with the default skin - that is the single biggest problem with it currently, IMHO. Now consider how a blind person using a screen reader or some handicapped individual using the OS accessibility features will feel about an application that not only does not let them use the application comfortably but does not let them use it at all, for what appears to be a completely useless feature from their standpoint ("We thought pretty pictures were more important than you being able to use the program."). - Performance - Mozilla's default skin is really slow; switching to the Sullivan skin literally made it go from slug-crawling-through-molasses to perceptually the same speed as native controls. Now, I appreciate that some new technologies (e.g. CSS) pose difficulties towards working with the system controls. Many of these problems could be alleviated by shipping a standard skin with each platform that uses either the standard controls or a close approximation and has been tuned for performance. Users would still be able to install custom skins but there'd be a lot less confusion among people who don't know/care about skins. The handicapped issue probably can't be worked around in the default skin, but I don't think it would be unreasonable to provide some alternate skin/extension option in the installer to use the standard controls. > but I don't think it would be unreasonable to provide some alternate skin/extension option in the installer to use the standard controls. And I don't think it would be unreasonable to keep NS6 complains to NS and Moz bugs to bugzilla. ;-) Sheesh, there's a simple solution sitting under all our noses and we're missing it. It seems that each OS would really like to have a skin that looks and feels like that OS. And this is actually really simple to do. Write the skins for each OS to work as that OS's users would expect. So you'll have a skin for Windows, a different one for MacOS (maybe even one for MacOS X) one for Linux (I'm happy with the current skin, but the 'Sullivan' skin looks quite nice too.), and you get the idea. When you download a specific version of Mozilla you get the skin that looks and feels like the OS. If you download the source you get the default skin. If you wan't to change your skin you can, BUT everyone gets what they expect when they first use Mozilla and that seems to be the reason behind all these remarks. And as a bonus, what a sales point. A product that works just the way you expect, but with all power to work the way you want it too! mintSlice Efforts to do what you suggested are on their way. It would be a simple solution assuming that everyone is using a "default" skin/widget/color... but the truth is that they don't (or there might not be a common or default skin). That is why I hope to see the CSS2 system color and CSS3 draft system font features implemented in Mozilla (See links from my earlier post). I thought I heard the GTK people were thinking about making it fully skinnable in the CSS sense, so native widgets would make sense. Does anyone know how far along this is? You know, I read the suck article and I thought it was rather funny. It was SATIRE people. Didn't you figure that out? I mean when he starts saying people shouldn't even be able to decorate their houses? Come on! As for this argument about Chrome. I personally think that XUL is 100% the way to go. I think that the Mozilla developers should make some standard looking "skins" that match the OS that they are running on for defaults. Simple looking skins. Those of us who like the freedom to customize will do so. I for one say the more free software is, the more we'll see software whose user interface is customizable. Go XUL, Go Mozilla, Go Linux, and Go Open Source! Have a great day! You have knowladge that I can only dream of. And you can sit here and argue over reasources, and if it's needed? Back up alittle and ask yourselves if Skins are what We The Poeple Want? I know that I do. I use IE because I cannot find a way to put my family photos on my tool bar. Hotbar lets me do that. But at the cost of being spyed on. Any and all browesers should have a way built in to add a toolbar skins. Most all of you have the ability to make that happen. But as I saw your all argueing over something that is totally irrelivent to the issue. I'm not as smart as most of you. But I see that Netscape and Mozilla are open source. We all have the right to make it what ever is needed. I'd really like to see Netscape 8.0 having the ability to add the photo of my grand child as the skin on the tool bar. It is the only reason that I use IE. And really to tell the truth I us Windows only be cause it is the only OS that anyone writes progams for almost. I don't care where you go you'll see that 90% of everything is made to run on it. The best software runs on windows. Give me an open sorce OS that well run windows programs. And I well be there and using it. Until then anyone that wants to try Linex is stuck with windows till you that has the knowladge can change it. We The Poeple Need YOU! Stop your fighting amonst eachother and work together to create what everyone needs. An opensorce OS that well run windows programs and not be pleaged by viruses. The virus creaters are after all after MicroSoft. Trying to drive them away from it. And MS is creating it to be vonerable to it also. On Purpus! I agree that MS has done wrong to free interprise. But most of us has not got the knowladge to chage it so we settle for what everyone tell us to. And one other thing that is needed is the ability to assign the swap drive to another drive. The internet browser swap as well as the OS swap. They fragment the OS drive more then anything else does. I have my win swap seperated now. But have not decovered the way to put IE's swap, ie temp on a seperate drive from windows. So this needs to be added into the OS as well as the browser. And any programs should be put on seprate drives or portitions also. I'm running win98. I have the swap on another drive. And it takes me about a half hour to defrag a 8 gig drive once every 2 weeks. And I never crash anymore unless someone gets through my firewall. But I also have everthing cloned on a seperate drive. So then I can just recopy it all to my C drive. Never do I have to warry about reloading drivers or nothing like setting up. Drivers are after all all thats loaded onto my OS's drive. Unless I change my drive configuration. Shortcuts you know. But I never have to argue with hardware, and software to get myself going now. What I'm saying is that all of this should be built into it all. And easy to set up for dummies like me. Screw the resources. Most everone that is buying new rigs are gitting more then a gig of memory. Well atleast for those of them that are buying 64 bit rigs.And I see no reason for even anyone to buy a 32 bit one with under 1gig of RAM. So the issue is not resorces. The issue is if your wanting to contribute to this cause. It's all up to you that have the knowage to help. For those of you that do want to help. Work together and make this happen. And above all stop your winning about it! Do something helpful for the world. Destroy companies like MS and get it done on thier turf. Thats whats gonna take them down. Not viruses. Just pure imagination. YOURS! Get about it we are all waiting and hoping for you all. Where is the new opensorce OS that well run anything thrown into it? Where's the Broweser that well let me dress it up as I want to? That well protect me while I'm on line? that well stop the spyware that is thrown at it, and viruses? You are the true defence that any of us has. Think about it. |