Slashdot Mirror
Dashboard Not a Konfabulator Rip-off
MacNN writes "John Gruber says the origins of Apple's Dashboard technology, announced as part of Mac OS X 10.4 Tiger earlier this week, are not with Arlo Rose's Konfabulator, but with Apple's original Desk Accessories and that Apple's Webcore-based implementation will allow many more developers/designers to create 'gadgets' much more easily and that Dashboard's 'gadgets' will offer much better performance: 'Dashboard is not a rip-off of Konfabulator. Yes, they are doing very much the same thing. But what it is that they're doing was not an original idea to Konfabulator. The scope of a 'widget' is very much the modern-day equivalent of a desk accessory.'"
I've seen a lot of arguments about the bits that apple 'copied' from konfabulator, and some are valid in that there's a lot of similarity between Konfabulator and Dashboard. Similar end function, similar look, similar workings underneath.
But it's all moot when you consider almost none of Konfabulator's implementation of the original desktop accessories concept was an original creation in itself. It might look like a big step to go from the 1984 desktop accessories to Konfabulator... and it is. but even THAT was done before It came after MS's built-in-to-windows Active Desktop, and after DesktopX, both Windows implementations of the same concept.
Good for Konfabulator for being a succesful product, but if Apple were to never use a concept that an external developer had previously used, then we'd have no desktop pictures, sticky menus, stickies, no glassy gui, no terminal, no dock, no onscreen clock, no login system, no web browser, no address book, no email application, no ichat, no full colour icons, no column view, no UI sounds, no font smoothing, no solid window dragging, no fontbook, no developer tools, no disk utility, no iphoto, itunes, sherlock, etc etc etc.
On the other hand, here is the official response from Arlo, the developer behind Konfabulator.
I think the most interesting point is that he knows about the 6-month head-start he has before Dashboard is available to the public.
What can be done in the mean time? Making more useful widgets. Porting to other platforms (the Windows port was announced in December). More importantly, enhance the application (maybe adding the same appear-only-with-key-pressed), and reduce the CPU load (using WebKit, it might also be possible to make Konfabulator Dashboard-compatible).
Surfin' Safari
He makes the excellent point that Dashboard/Konfabulator-type of widgets have been done in browsers, too.
My comment about Watson/Sherlock stills seems applicable: don't whine, give us a better product.
"All the darkness in the world can not quench the light of one small candle."
Internet Explorer 6 only gets the box model wrong if you kick it into "quirks mode".
Ditto, except I use Gecko as my reference rendering, and that usually translates almost seamlessly to KHTML and Opera, with large amounts of messing about to get a decent rendering in Internet Explorer.
To quote from the site:
Basically, the whole Widgets/Konfabulator issue is a load of horse manure cooked up by Arlo Rose to generate some publicity.
To celebrate the occasion of my 1000th post, I will post no more forever on Slashdot. Goodbye.
Yeah just have to chime in again here, IE's implementation of CSS is so much significantly worse than Safari, Moz & Opera that I consider it a totally separate stage of developing my sites. First I go through and develop for the browsers that agree on 95% of how to render the CSS.... then IE gets at least twice as much time of bug fixing in order to identify how it's going to screw up this time. The problem with IE that I see being the largest is its ubiquitous nature, and the fat that it was so friendly towards bad html coding. A lot of people got used to testing sites in it, because - hey - most of the time they'd work great. Then when these people started switching over to more standards-based design practices they saw how screwed up everything got. Using IE as the basis for reference it's easy to blame the others. The only problem with that is IE ignores SO many aspects of the CSS standards that it really is apples v. oranges. Zeldman talks about it in his book some - Designing with Web Standards - one of my favorite sections is about the infamous box model problem - it's great, because to most uninitiated, the original IE broken box model actually makes much more sense... but its not the rules :/
I'm not sure this is correct - looking at sites in IE 5/5.5/6.0, they all seem to implement the box model identically even when I provide them with a complete DOCTYPE, which *should* ensure non-quirks mode.
And, I do still have to support IE back to 5+. it would have been nice for them to have gotten it right the first time around. It's actually not that much of a problem now that I can anticipate it, but still a hassle to have to build in multiple-nested divs and not be able to rely on a given behavior.
Are you mad? Safari's CSS support is second only to Mozilla, as is their Javascript/DOM support. I know this because I've built an in browser, live CSS editor and tested it extensively across most all the various browsers.
I can even name areas where Safari's DOM handling surpasses Mozilla (although I can name more where it doesn't), and a few CSS instances where Safari beats Mozilla too (although again, more where Mozilla wins).
Opera is up there in CSS, but falls down in some DOM areas, and IE isn't in the race at all. To claim that *any* browser is "far more CSS compliant than Safari" is stretching the truth well beyond breaking point, and the only one you can probably honestly claim that does beat it is Mozilla.
I disagree. Safari's CSS support is not second to Mozilla's; Safari implements quite a few things which Mozilla doesn't (for example, text-shadow), while I don't know of anything which Mozilla implements which Safari doesn't. Plus, as I posted elsewhere in this thread, Mozilla has a few bugs with background-attach:fixed (though at least it supports it, unlike IE which only supports it on BODY).
There are 10 kinds of people: ones who understand ternary, ones who don't, and ones who think this joke is about binary
The current Konfabulator home page more than makes up for that. ("Cupertino, start your photocopiers")
Ok, I'll bite... Dashboard is neat as much as Konfabulator; I'm just wondering how hard on resources it is. You see, when Konf. hit /. I sheepishly proceeded to download it, ohh it, check out the cool widgets! Then I got bored, many widgets bore the disclaimer "hey, I wrote it off a boring weekend!" and noticed the fan humming... I have a laptop, I don't care for a translucent gooey can showing off how fast it's draining my battery. So if Apple's stuff is fast and chews less juice I'm all for it; (sarcasm) had it been really original they sould have patented it right!? (end sarcasm)
On a side note... everyone here must be missing the real people getting pissed off: Adobe. What will they do to justify Photoshop's price tag now that Core Graphics will make writing very close software as easy as a shareware tutorial!
Mi domando chi à il mandante di tutte le cazzate che faccio - Altan
It's pretty hard. I used to have 4 windgets running: time zone widget, CD cover widget, lyrics widget and weather widget. Then, I noticed performance hits and when I checked Process Viewers, I found out that each widget run as a separate process, each with CPU and memory allocations on the top of the application itself. Also, there seems to be a problem with memory leak as the performance got worse overtime, which might be due to bad JavaScript/AppleScript code. IMHO, Konfabulator should provide better code checking.
Anyway, now I don't have any widget running all the time. I launch widgets when I need/want them.
But when Apple crushes all the Arlo Roses of the world, who's going to be left to write software for our precious Macs?
I think you are confusing utility and app developers. Utility developers by definition are filling in minor holes in the operating system which they should expect to be filled soon by the actual OS. Quaterdeck made a killing on memory management when DOS & Windows needed it but didn't have it (or only had a bad version). Norton made a killing when disk defraging wasn't included with Dos, etc... So things like Watson and Konfabulator are to be expected.
Conversely things like Final Cut Pro and IE was just Adobe / Microsoft chickening out. Apple was just bringing out an app. Omni for example is OK with competing with Apple (and they've got 2 of their apps actually bundled in with the professional lines). In any case Apple writes great software and more dependency on Unix apps (Linux) would be great for both sides. Apple is the expect at exactly those areas where Linux developers tend to be lacking and vice versa.
I tried using Konfabulator. At first I thought it was great, then I realized that the mini digital clock was using 60% of my CPU! It made my computer run like crap. Now, there were other widgets that ran without problems and used tiny fractions of the CPU cycles. So I turned of the resource hogs. Then I started thinking about security and decided that since these things could be written by anyone, what might be happening? So, maybe in a paranoid, naive, uninformed decision, I quit using Konfabulator. I also didn't want to pay the fee.
I'm thinking that Apple's architecture for thier widgets would incorporate the security measures already in web based media (for whatever that is worth) The author doesn't really bring up security, so I'm still wondering if it's possible to create a naughty Konfabulator widget that looks like some innocuous tool, but is actually doing bad things. Or could it be done in Apple's new model?
I'm looking forward to it, though.
Slashdot Eds Link Anonymous Posts With Logged Posts
They Are Vermin Feeding On Each Other's Feces.
I Hate \.
I agree. I also used to have Konfabulator installed, and found it a bit unwieldy. (Admittedly, when I had it installed, it was on a 350Mhz G3, so it was slow.) Taking up space all the time just got in the way.
So when I heard about Dashboard, my first thought was "Another Konfabulator. Cool idea, but not that useful in the long run." It wasn't until I saw the Exposé-like disappearing act that I realized how much it was really needed.
I missed Desk Accessories. I use Stickies and Calculator alot. I used them alot in the Classic Mac OS. (Going all the way back to System 5, when Stickies were 'Note Pad'.) I liked the quick access, quick loading, and state-saving of them. Yeah, having Stickies and Calculator in the dock is what I'd ended up doing. (On my notebook. On my desktop, I use a Kensington mouse with a 'quick-launch' button map, so I have Stickies, Calc, and Terminal in the quick-launch menu.) But even though I use them often, it's a bit cluttering of the Dock. (Which I like to save for those programs that I really do load 'all the time', to keep a de-cluttered appearance.)
So the new Exposé-like effect of Dashboard just seems to be the perfect implementation. It combines the 'small widget' of Konfabulator with the ease of constant access of Desk Accessories, in a good way. I can't wait for Tiger. (When Panther came out, it took some convincing to see how good Exposé was. With this, it's obvious.)
Another non-functioning site was "uncertainty.microsoft.com."
The purpose of that site was not known.
However, in current versions of Mozilla, if you add an image background with background-attach:fixed to the second div, the meta-lorem-ipsum will still continue to wrap around where the first div's content should be, but the second div's background will paint over the "blah"s. This is obviously a Mozilla bug.
There are 10 kinds of people: ones who understand ternary, ones who don't, and ones who think this joke is about binary