April 24, 2003

O'Reilly Mac OS X Innovators Contest Winners

Congratulations to Brent Simmons (NetNewsWire) and Robb Beal (Spring) for their wins in the O'Reilly Mac OS X Innovators Contest. I'm happy to be a user of both pieces of winning software and am looking forward to seeing what comes in the future.

Posted by kstaken at 12:04 PM | TrackBack

April 23, 2003

Edgar Codd, database theorist, dies at 79

The mathematician laid the theoretical foundation for the standard method by which information is organized in and retrieved from computers.
The New York Times [CNET News.com]

Posted by kstaken at 12:32 PM | TrackBack

Using Spring to Build a Better eMusic Stash

I'm a huge fan of eMusic and recently they introduced a new feature to the service called My Stash. This allows you to track albums that you want to download in the future. For heavy eMusic users it's very useful, but being a web based app the user experience isn't what I'd like. I'm definitely a big believer in the beyond the browser idea and being the eMusic nut that I am, I wanted a better solution.

I found most of what I wanted in the Spring Desktop which I already had a license for. Unfortunately Spring is early in its life and is pretty immature right now. In particular there's no way to have it create custom objects from dragged URLs. After talking with Robb Beal (the creator of Spring) about this. I set out to find another way to drag links from the browser, create spring objects from them and then place them on a canvas. I knew it would be easy to do this as a multi-step process, but the usability of that kind of solution, well to put it simply, stinks. I wanted to be able to drag from the browser and into Spring in one motion so I came up with what I'm calling the "Magic Wand" approach.

I created a small Cocoa app that sits innocently in the corner and doesn't do anything until you drag something over it. When you do this it looks at what's being dragged and sees if it can create a spring object from it. If it can it creates the object and then swaps the pasteboard contents so that when you continue your drag and drop it on a Spring canvas the object gets added automatically. Pretty cool and works great. So now I can drag album links from eMusic and easily create Spring objects from them that enable some really useful functionality. Here's a screen shot of what you can do with the objects once they're created.

Overall a very handy application and for eMusic fans at least, a great use for Spring. If you're an eMusic subscriber and Mac OS X user it might be worth checking out. Spring is commercial software, but a license is fairly cheap at $21.95 and there's potential to use it for more then just a better eMusic stash.

Here's the download for the Album Magic Wand application. This requires Spring, and if you want to download you have to have an eMusic subscription and a copy of Pickup configured to manage your downloads.

Posted by kstaken at 01:30 AM | TrackBack

April 22, 2003

New programming guide to be available soon

[MacNN]

Hadn't heard of this book before, but it sounds interesting. There's definitely a need for more advanced books on programming for Mac OS X.

Posted by kstaken at 10:41 PM | TrackBack

Deep inside Apple's Piles

Forgotten UI innovation to be exhumed? Inventor speaks to The Reg [The Register]

While piles are interesting, to me the really interesting rumor floating around is about a new file system. This is something that is desperately needed. BeOS was the first step, but there's much that can be done beyond that. Current file systems are severely holding back the usability of computers. One of these days I'll actually try to write up why I think this is the case.

Posted by kstaken at 10:38 PM | TrackBack

April 21, 2003

A music related special event with Apple on April 28

via MacCentral: Apple has several music products, but speculation over the past couple of weeks has focused on three things: a new iPod; a buyout by Apple of Universal Music; and a new music service by Apple. [Mac Net Journal]

OOH, I sure hope this is more then just updated iPods.

Posted by kstaken at 02:24 PM | TrackBack

Another Cocoa Word Processor

Just came across another Cocoa word processor called Mellel. Looks like it's been around for a while, but I'd never seen it before. It falls into the abuse the metal interface catagory and is in clear violation of Apple's UI guidelines in this regard (but so is Safari, so who knows what the guidelines are now). However, regardless of what the guidelines say this interface actually looks pretty slick. The program seems to work a little different then the norm, but that's definitely not a bad thing if it can also work better. I still don't have any real need for a word processor, but this looks like another decent alternative for a basic solution. The price is only $19 which is a real bargain compared to many of the alternatives.

One major complaint, the file format is binary with an embedded property list. This is the age of XML, and any modern program should either have an easy way to store it's data in XML or better, simply use XML for its native file format. I'm extremely hesitant to rely on any software where I can't understand the file format from the perspective of easily repurposing the data. It seems Apple is starting to get this idea (i.e. iTunes and iPhoto making XML duplicates of their DBs and the Keynote file format), it's time that other people developing for the Mac did too. This particular program can export RTF and plain text, but no XML. Of course Apple could make things easier in this regard by providing something that resembled decent XML support in the Cocoa frameworks.

Posted by kstaken at 04:03 AM | TrackBack

NeXT still stands out in its Mac incarnation

Dan Gillmor: “Software developers loved what they could create with the NeXT platform. And for many, OS X has meant a phoenix-like rebirth.” [ranchero.com]

Posted by kstaken at 12:02 AM | TrackBack

April 17, 2003

New XML:DB API Implementation

Lars Martin just posted an announcement from Cincom saying that they've implemented the XML:DB API. They're claiming that they support Core and Transaction which is kind of amusing since we never actually finished Transaction, but it's all good I guess. I haven't worked on the XML:DB API in a long time and the project is badly in need of new blood to continue the work. There are many, many things that could be improved greatly with it.

Posted by kstaken at 05:37 AM | TrackBack

Microsoft Wants Your Love

Today, Don Box of Microsoft
appeals
for love
(well, at least polygamy), inviting us all to snuggle up with
their .NET stuff and pointing out that it works with Open Source technology.
Scoble takes a Redmond job and also wants to
start
dating.

I have this problem with loving Microsoft, it's just that I
fear them so much.... [ongoing]

Tim Bray on his fear of Microsoft, a feeling I know all too well.

Posted by kstaken at 04:15 AM | TrackBack

April 16, 2003

Transparency

God help us if we're wrong. [Don Box's Spoutlet]

Don's talking about the increasing number of Microsoft employees who are blogging and the increased transparency it's bringing to the company. This is a trend that I'm very impressed with and is providing a fascinating look at some of the smart people who work there. It's also bringing a better level of community and personality to what had previously been a very cold company. I've been adding more and more Microsoft related blogs to my regular reading and it's definitely improved my feelings about the .Net era Microsoft. Even if the amount of transparency is currently small, it sure is better then what existed before and it seems more Microsoft bloggers are coming online everyday. Of course it will take a lot more then this to regain trust and things like WPA, Palladium and the completely broken mess that is Windows don't help any, but it's good to see things improving and there are definitely some things in .Net that I really like.

I'm hoping that we can see more blogging in the Macintosh developer community. Especially among Apple employees, but also among independent developers. I'm only aware of two Apple employee blogs, Ken Bereskin and Dave Hyatt. Both are great resources, hopefully there are more out there that I just haven't found yet. Dave Hyatt's blog in particular is interesting because of the development oriented discussion around Safari. Given Apple's extreme stance on secrecy it has to be a tough line to walk, but I sure wish more Apple employees would jump in.

There are quite a few independent Macintosh developers with blogs, many of which were very early adopters. People/Companies like Brent Simmons / Ranchero Software , Don Watson / Karelia, Robb Beal / UserCreations, Michael McCracken, Unsanity and CocoaTech and of course I'm working on Mac software now too. I haven't publicly released any Mac OS X software yet, but it will come someday.

Macromedia is another company making big strides in this area with people like Mike Chambers, Kevin Lynch and Jeremy Allaire blogging. The blogs offer an interesting mix of company marketing and personal insight. Hopefully more of the latter, but regardless it puts a much more human face on the company. In fact, I don't actually use any Macromedia software, but still read some of the blogs just to keep an eye on things. Actually, Macromedia Central is something that I am interested in, but for now I'm just watching it.

Posted by kstaken at 10:48 PM | TrackBack

April 14, 2003

Safari Public Beta 2 released

Apple finally released a new public beta version of its Safari Web browser this morning, bringing tabbed browsing, better standards compatibility, improved AppleScript support and autofill forms and passwords. [Mac Net Journal]

Good to see an official release with tab support.

Posted by kstaken at 12:35 PM | TrackBack

Some handy Cocoa source code

I'm pleased to announce the opening of an open-source repository of some useful Cocoa tidbits used in Watson and some other projects I've been working on. Unlike the Omni Frameworks or CocoaTech's new frameworks, this is a lot less ambitious and less packaged. It's really just a weblog (Yes, Virginia, there is an RSS feed) where each entry is just a snippet of code. You need to copy and paste into your projects as needed. The copyright is, I believe, extremely liberal (certainly not infectious like the gnu license), but let me know if you feel it can be improved.


Why do this? Just a way of giving back to the Cocoa community that has been so helpful to my efforts in building Watson.


I'll add to this repository occasionally as I run across more generic bits of code, but don't expect a steady stream! :-) [Karelia Software ]

Excellent, we need more of this.

Posted by kstaken at 12:32 PM | TrackBack

April 10, 2003

ShareKMC for Tablet PC

Another interesting toy from the Microsoft world. There's no reason this kind of thing couldn't work on a Mac, or even between a Mac and a PC. It would be a lot nicer then using a KVM switch.

Posted by kstaken at 07:55 PM | TrackBack

XPath Queries over Objects

A very interesting article on MSDN about querying objects via XPath. This is fascinating stuff and really brings home how vastly superior .Net is for XML when compared to what's available in the Cocoa frameworks. In fact Apple's XML support is so pathetic I'm embarrassed to even talk about it.

Come on Apple, XML is the most important technology around and the Cocoa frameworks have virtually no support for anything of value in the XML arena. The pathetic Core Foundation XML parser is a joke. It's a C API, doesn't directly support namespaces, has no DOM, no SAX hell no anything that anyone's ever seen before. Forget about getting XPath or XSL-T working with it. Garbage, garbage, garbage.

Sure we can drop out to Java, or C or C++ or even Perl and Python, but why is this necessary? Objective C is a perfect language for building awesome XML tools and solid XML support is critical for building network enabled applications. So far Apple's really dropped the ball in this arena. I'd say third parties should build better tools, but XML is truly foundation now and solid support absolutely must be baked into the basic platform. Apple's has some interesting stuff, i.e. the XQuery implementation that's part of Sherlock, it's just locked up in places that are useless to other app developers. It could be so good, but instead it's so sad. I'm getting .Net envy and I'm not happy about that.

Posted by kstaken at 04:02 AM | TrackBack

BBC: Concorde grounded for good ....

BBC: Concorde grounded for good. [Scripting News]

Wow, what a bummer. The Concorde was cool even if impractical. 30 years ain't too bad a run though.

Posted by kstaken at 03:27 AM | TrackBack

This could be the year for UDDI

This could be the year for UDDI
[InfoWorld]

Hah, hah, yeah right.... UDDI is dead, dead I tell you. Taking an already overly complex, useless specification and making it even more complex just isn't going to solve the problem.

Posted by kstaken at 01:52 AM | TrackBack

OmniWeb 4.2 released, 4.5 details revealed.

MacCentral: OmniWeb 4.2 released, 4.5 details revealed. [Hack the Planet]

4.5 is the first version that's using the WebCore and JavaScriptCore libraries from Safari. Should be very interesting to watch how this develops. I know I'm certainly excited about the prospect of having a first class HTML rendering component available for application development. Of course it's also sad that this is even interesting, Windows has had the capability for a long time and it was something that in that context was considered an evil move by Microsoft. Hypocrisy is great, oh well, bring on the WebCore based apps.

Posted by kstaken at 01:45 AM | TrackBack

April 08, 2003

Rethinking Digital Music Services

Lately I've spent a bit of time looking at the various legal music services. In particular listen.com Rhapsody, Pressplay and eMusic (which I subscribe to). I've posted several times that I'm a big fan of eMusic and it's a model that I'd love to see replicated with services that have access to the catalogs of the majors. However, from a business perspective I also have to say I don't see that as a practical idea.

I've never been a fan of services like Pressplay. This is mainly because it just feels like an artificially hindered service compared to what is possible (which it is). The restriction to WMA files that have to call home once per month to renew their licenses is just annoying. However, given their new pricing model this isn't really a terrible thing. For $9.95 per month you get unlimited downloads. What makes this model still seem undesirable is that all those downloads become useless if you cancel the service or if the service goes under. The question is why is this a problem? It's all a matter of perception. I believe they're actually shooting them self in the foot by focusing on the idea of downloads.

The other service I've been looking at lately is listen.com Rhapsody. This service doesn't offer downloads at all. It's more a music on demand streaming service. This means you need a connection to access the music. They do allow you to pay extra to burn tracks to CD, but do not offer any way to download a portable music file any other way. Now even though this is technically more limiting then Pressplay, I find the service more appealing. Why is this? I think it's because I feel I like Pressplay is trying to misrepresent what it is they're offering. They aren't really offering downloads, they're offering streaming music on demand service that just happens to cache on your hard drive in WMA format. This just feels dishonest and smacks of trying to pull the wool over the eyes of the consumer. In contrast the Rhapsody service doesn't claim to be anything other the a streaming service and at $9.95 per month for unlimited streams it's actually not a bad model. You just have to compare it to services like satellite radio to understand the value. XM radio is almost the same price (more if you factor in hardware) and you have no control over when and what music you hear. With listen.com you have access to a large catalog and can choose exactly what you listen to.

For me, when I think about why I like digital music so much, it's all about convenience and quantity. I like being able to call up just about any music that I get curious to hear with just a few key strokes. I've always wanted access to any kind of music, it's why I own so many CDs. Now if I can have access to huge amounts of music without having to worry about storing all of it, why wouldn't I go for that? So to me listen.com Rhapsody actually looks like a really interesting service. I haven't tried it out because it's Windows only, but someday when something like this comes to the Mac I'll be waiting. I'm changing my feeling about these services. I don't really care now if they offer unlimited downloads, all they really need to offer is unlimited streaming and a HUGE catalog that extends deep into their vaults. If I get curious to know what some obscure blues artist from 1922 sounds like I should be able to just call it up and listen. I don't need the MP3 file for that, a stream is fine (as long as it's high bit rate). That's a service I'd absolutely love.

So here's my new wish list in a legal music service.
- Complete access to the full catalog of as many major and independent labels as possible. I mean everything. If an album is there, the whole thing should be there.
- Solid well designed and easy to use software for Mac OS X
- Unlimited streaming of at least 128kbps.
- $9.95 per month
- The ability to purchase unhindered MP3s of songs for $.20 per song. (I doubt this will happen, but it'd be nice and isn't a deal breaker)
- No ads, no cross promotion of partners crap, no SPAM or anything else that distracts from the purpose of the service.
- Not marketed as a download service.

The last point is the one that will be most difficult to get people used to P2P to accept. These services need to focus their marketing on their strengths and downloading songs isn't one of them. Convenience can be though, and for the people who are actually willing to pay for music anyway, it's worth paying for convenience. So now listen.com comes close to what I want, and I'm hoping the rumors about Apple coming out with a service are true.

Posted by kstaken at 08:42 PM | TrackBack

April 07, 2003

Genghis Blues

First it was Kronos Quartet recording a classical set with the throat singers of tuva, now I just discovered a blues recording with Paul Pena and some tuvan singers. eMusic is great for finding fascinating music like this. It even includes a blues styled version of Kongurey which is the song Kronos recorded. A tune I've always really enjoyed. Fascinating stuff, the tuva singers are like nothing else. Great stuff if you have an open mind.

Posted by kstaken at 04:40 AM | TrackBack

April 06, 2003

Introducing SharpReader

Luke Hutteman: handles all RSS versions, modules like dublin core,
content:encoding, xhtml:body, etc. ... RSS Auto-discovery ... and last
but not least, for items that include the full html description,
sharpreader lets you expand a headline to view links and responses
to/from other posts in other feeds. This allows you to read posts in
context, and will show related posts together. [via Phil Ringnalda] I
really like the related posts feature. [Sam Ruby]

This actually looks pretty interesting. I've been reading a lot of .Net blogs lately and the community there seems to be picking up quite a bit. In particular there are a large number of Microsoft employees blogging now. Something I wish Apple would encourage more of.

Posted by kstaken at 06:09 AM | TrackBack

April 04, 2003

Why is Applescript so hard

Applescript is a strange language. It is easily one of the most readable programming languages I've ever seen. Yet, no matter how many Applescripts I read, every time I go to write one I run smack into a wall of befuddlement. I've used a couple different books on it and I guess it's just the complete lack of documentation that exists for the actual application implementations that is so problematic.

It seems app developers are intent on relying on the scripting dictionary for documentation along with maybe a couple sample scripts. This is just not adequate. I'm constantly amazed at what's possible with Applescript and even more amazed that the people who write these scripts were actually able to figure out how to do it. This is especially frustrating to me as I know dozens of programming languages and have used 10-12 different languages professionally, but writing Applescript is just befuddling.

It's like the complete opposite of Perl. With Perl once you learn the language programs kind of just flow out of your mind and into the computer in a very expressive fashion. However, reading Perl code, especially code written by people who take pride in using all the Perl shortcuts, can be a quite painful experience. Applescript is the opposite, reading the code is always easy. I've yet to see a script that I couldn't understand. Writing it though. Ugh, why is it so hard? It just doesn't make any sense.

Obviously, I've been doing a lot of work with Applescript or rather more in using Applescript with Cocoa applications. I'm excited by the capabilities, but frustrated with how difficult it is anytime I have to step into the Applescript domain and write a script from scratch. It's still fun though.

Posted by kstaken at 05:52 PM | TrackBack

PowerBook Rumors (17 and 15inch)

Powerbook revisions... [MacRumors]

Now that's a cool rumor. A dual processor 17" powerbook would be sweeeet!

Of course actually having the dual proc Powermac, that I already own, actually working would be sweet too. The Apple store has had the thing for over a week, I'm getting a little pissed off about this.

Posted by kstaken at 04:55 PM | TrackBack

April 03, 2003

Free content: Why not?

Venture capitalist Greg Blonder says there are hard-headed business reasons that support the notion of giving away music. [CNET News.com]

An interesting idea. The current music distribution model is definitely doomed. MP3s are just too convenient. Services like eMusic show that wide availability of large collections of music is an outstanding value.

File sharing networks are a waste of time. If I could spend $20-$30 maybe even $40 per month for unlimited downloads of unrestricted MP3s from ANY label I'd jump on that in 2 seconds. And guess what, that's $20-$30 more then the record companies are getting from me now. I was one of their best consumers, 2500 CDs worth proves that, but I'm not buying anymore. They've priced them self out of the market and their greed just makes me sick. I remember the days when CDs were new and tapes and records cost $5-$8 while CDs were $12-$15. I remember the promises that it was just the cost of new technology. Well, bah, I can produce a CD on my computer for less then a dollar it can't possibly cost them more then 10 cents in bulk. So where's the price reduction? Heh, price increases are all they've been doing. New releases are now $19 for some labels at full retail.

What really sucks is that they're using this artificially inflated pricing to determine value for their online services. Basically trying to make the price of the files purchased online to be equivalent to the price of an inflated CD. Yeah, right. Here's the deal $.99 per track is too expensive. Want to create a killer service? Flat rate all you can eat pricing and unencumbered MP3 files. Get major label music on there and music consumers will line up to buy it. The rest who actually have the time to waste on file sharing networks? Forget about them, they're not buying music anyway. A solid easy to use service is worth paying for.

If CD sales are dropping maybe they should look a little more at their value proposition. They're actually punishing the people who like to buy music. I like to buy my music, I believe the artists should certainly be well compensated. The record companies though, well I certainly won't be heartbroken if they disappear. For the time being I'm sticking with my current CD collection and enjoying greatly what eMusic has to offer. If prices on CDs come back to earth, maybe I might even buy a few again.

Posted by kstaken at 08:33 AM | TrackBack

April 01, 2003

Independent Macintosh Shareware Developers Group

John Kinsella is exploring the idea of creating an Independent Macintosh Shareware Developers Group. Seems like an interesting idea, but it's something that needs to be explored in an open forum. I believe there are/were organizations that did this before.

Posted by kstaken at 06:38 PM | TrackBack