Richard Allan Baruz: Computationally Minded

Fri, 15 Nov 2002 07:18:31 GMT

[The Creation of a Constraint-based Programming Language]. I had to paraphrase the title to fit the maximum length, it's really: The Definition and Implementation of a Computer Programming Language Based on Constraints, by Guy L. Steele (thesis) This is a really wonderful piece of writing. What I got out of it was an introduction to constraint-based programming, a gentle tour of growing a language bit by bit, and an amazing demonstration of Lisp techniques for writing interpreters that fit in neatly with their host environment. (And I only read the first half!) That said, it is book-length and the electronic copy is a scan. The result is 20 megabytes and challenging to print - but well worth the effort. [Lambda the Ultimate]

Steele Jr

Mon, 04 Nov 2002 15:10:41 GMT

I thought I posted this before. Oh, well. Paolo describes how to host a Radio site on a .mac account. This could probably be applied to any WebDAV implementation.

Mon, 04 Nov 2002 01:57:28 GMT

By the way, today is the one-year anniversary of the TiBook. Er, my TiBook. I know because I am struggling to register its APP before the end of the one-year cutoff. No go. :.( On a brighter note, OmniWeb seems to be picking up the Userland control for text editing, which Navigator does not. On a not-so-bright note, OW is still much slower than Chimera.

Sun, 03 Nov 2002 16:01:47 GMT

Benchmarking these Smalltalk Luhn implementations: a _ [ :x | x isValidLuhn] . b _ [ :x | x isValidLuhn2]. interval _ (50000000000 to: 50000010000). Time millisecondsToRun: [interval select: a] 18071 Time millisecondsToRun: [interval select: b] 28953 After tinkering around with both algorithms, I realized that keeping a running total is much faster. An even greater speed improvement is achieved by converting the ASCII character to its corresponding integer then subtracting 48, rather than converting to a string then converting the string to a number. So instead of aDigit asString asNumber, doing aDigit asInteger - 48 does better: Time millisecondsToRun: [interval select: f] 4768 The naive implementation is easier to read, but three to four times less efficient, as it conses up a string from a character. Using (aDigit asInteger - $0 asInteger) might be a slightly more readable compromise. It seems like using a dynamically allocated Collection of any sort seems to be very bad for performance. Combining the mapped doubled add-the-digits approach with the accumulator approach: Integer>>isValidLuhn8

	| stream accOdd accEven anEven doubleMap |
	accOdd _ accEven _ 0.
	stream _ ReadStream on: self asString reverse.
	doubleMap _ #(0 2 4 6 8 1 3 5 7 9 ).
	[stream atEnd]
		whileFalse: [accOdd _ accOdd + stream next asInteger - $0 asInteger.
			anEven _ stream next.
			anEven
				ifNotNil: [accEven _ accEven
								+ (doubleMap at: anEven asInteger - $0 asInteger + 1)]].
	^ accOdd + accEven \ 10 = 0

...seems to yield really good performance (3.5 to 5.5 seconds for 10001 operations) on average, while being only slightly incomprehensible.

Sun, 03 Nov 2002 05:41:32 GMT

I realized that since credit card numbers are valid Luhn numbers, I could test my algorithm against them. It works for my credit and debit cards, as well as my Waldenbooks Preferred Reader card (from the eighties), but not for my Barnes and Noble card number or the Replay card. I tested the implementation against those, then another: Integer>>isValidLuhn2

	| str accOdd accEven anEven |
	str := ReadStream on: self asString reverse.
	accOdd := Bag new.
	accEven := Bag new.
	[str atEnd]
		whileFalse: [accOdd add: str next asString asNumber.
			anEven := str next.
			anEven
				ifNotNil: [accEven add: anEven asString asNumber]].
	^ accOdd sum
		+ (accEven
				inject: 0
				into: [:x :y | (#(0 2 4 6 8 1 3 5 7 9 ) at: y + 1)
						+ x]) \ 10 = 0

Sun, 03 Nov 2002 03:06:50 GMT

I tried my hand at coding the Luhn algorithm in Smalltalk, as almost every other language seems to have had a shot at it. I am certain that are better implementations than this quick hack, but here it is. It works true for the example case that the accordian guy uses, but having no other test cases, this is what I have. Note that this is in the Squeak dialect, and it attaches to a base class. Integer>>isValidLuhn

	| str accOdd accEven anEven |
	accOdd _ accEven _ 0.
	str _ ReadStream on: self asString reverse.
	[str atEnd]
		whileFalse: [accOdd _ accOdd + str next asString asNumber.
			anEven _ str next.
			anEven
				ifNotNil: [anEven _ anEven asString asNumber * 2.
					anEven > 9
						ifTrue: [anEven _ anEven - 9].
					accEven _ accEven + anEven]].
	^ accOdd + accEven \ 10 = 0

[update: Whoops. I was using a stripped image (the image I use for my Comanche Swiki), so all the variable names were lost, so I replaced them. The underscore is the keyboard key used for the assignment operator, which in Squeak images renders as a left-pointing arrow. The caret sign is an up arrow and returns from the function. update: There must be a better implementation, probably using #inject:into:. The implementation above streams through the digits of the integer, accumulating the odd and even digits alternately. update: I don’t know why the algorithm description insists on adding the digits of the doubled even-positioned digits; it seems to me that the digit-added property holds true even if you add the doubled digits together after, instead of at each step. That is, 35 + 11 and 8 + 2 both ultimately equal one. I don’t remember what this property is called. If that is the case, the implementation above could be simplified to something like, um, oh, wait, never mind. Sequence of actions is important. Never mind. update: I originally pasted in the file-in version, but I thought if I presented as it is presented in Kent Beck, it would be less scary. Smalltalk really is a pretty language. update: Whoops, more Smalltalk implementations just popped into my head: here and here.]

Sun, 03 Nov 2002 02:21:47 GMT

An oldie, but Paul Graham writes for the ages. Good design...

...is simple
...is timeless
...solves the right problem
...is suggestive
...is often slightly funny
...is hard
...looks easy
...uses symmetry
...resembles nature
...is redesign
...can copy
...is often strange
...happens in chunks
...is often daring

Tue, 22 Oct 2002 05:33:51 GMT

Wow! A really technical discussion of the Mach-O ABI in Mac OS X on...wait for it...the Slashdot Apple site! Done in response to a posting on the Unsanity.org blog.

Mon, 14 Oct 2002 22:20:25 GMT

Another implementation of the SGI GL library, WOGL.

Mon, 14 Oct 2002 00:44:23 GMT

Noodling about. David Gelernter, the mind behind Linda, has an article at CIO Insight. In the Expert Insight column, an interview with him: Rethinking the GUI for the Big Picture. Gelernter has been pursuing this vision of a new interface for some time, and has his own company to pursue these ideas. (seen it mentioned Comm of the ACM and Nooface) He treats lightly with the idea of communities of practice, a term I am only slightly familiar with in its corporate context. I think it may have arisen from postmodernist theory; I think Stanley Fish speaks on the idea of communities of practice, and how an idea of truth arises from these communities by way of a shared understanding. (Remember, postmodernists care little for the idea of an objective truth.) The idea of streams of information, creating a narrative, has other proponents, such as Ted Nelson and his famous holy grail, the Xanadu project. I believe with ZigZag, Ted Nelson wants to be able to create several streams across the same data, connecting them in different streams to present them in different ways, such as a chronological stream for project history, a criticality stream for project priorities, and, say, several substreams for individuals’ responsibilities in a project. Scopeware seems to me to be Gelernter’s attempt to unify the lifestreams approach with several other interface ideas he’s presented before. [by way of MacCentral]

Sun, 13 Oct 2002 16:15:36 GMT

uptime | cut -d' ' -f3-7 | sed -e 's/.*/say "&"/' | osascript & Uptime: the Mac OS X killer app (Doc Searls?), but speakable, thanks to Mac OS X Hints. Going to restart soon, I guess. [rizal:~] baruz% uptime 11:14AM up 16 days, 13:56, 3 users, load averages: 4.29, 4.37, 4.14

Sat, 12 Oct 2002 01:05:45 GMT

The redoubtable Andrew Plotkin likes Rhem. Best of all, Mac support! Bad news: I don’t have time.

Tue, 08 Oct 2002 21:42:24 GMT

Accessorizer / Kevin Callahan. (SOURCE:Ranchero)
<quote>
Accessorizer is a FREEWARE application and a set of Services which generates ObjC accessor declaration and implementation methods from variable declarations. Accessorizer offers a broad range of memory management schemes used by the best-known Cocoa/Objective-C experts. The readme includes links to important memory management articles and resources.
</quote> [by way of Roland Tanglao's Weblog]

Tue, 08 Oct 2002 21:38:51 GMT

Tecplot arrives for Mac platform [MacCentral] More information $1495 :./ If only...

Sun, 06 Oct 2002 16:59:36 GMT

steal your interface

Sun, 06 Oct 2002 06:59:11 GMT

“Why I Take Good Care of My Macintosh,” by Gary Snyder. I have this recording somewhere, from the In Their Own Voices boxed set of poets reading their poetry.

Sun, 06 Oct 2002 01:28:33 GMT

Rendezvous (zeroconf) talk notes

Sun, 06 Oct 2002 01:12:07 GMT

Lambda the Ultima’s research papers page.

Thu, 03 Oct 2002 22:32:05 GMT

One reason I block image download in Mail.app.

Porn Spam: It's Getting Raunchier. Disturbing, explicit and sometimes-illegal pornographic images are popping up more and more, unsolicited, in e-mail boxes everywhere. Observers worry even the sickest perversions may become more mainstream as a result. By Julia Scheeres. [Wired News]

The other: I don’t want spamsters knowing I even exist by way of cookies in embedded images.

Mon, 30 Sep 2002 21:08:05 GMT

Bletcherous URLs. One of the things I love about the original WikiWiki was that you could look up the thoughts of the community by simply typing a probable title in the CGI query string in the accepted Smalltalk-like convention (i.e., http://c2.com/cgi/wiki?YouAintGonnaNeedIt). The Comanche Swiki uses numbers as a primary lookup, but you can use probable URLs that will resolve to a close match, which is nice, but may have problems if trying to refer someone to a specific source without the URL on hand.

Fri, 27 Sep 2002 17:47:07 GMT

Cleaning up my work inbox (our mail server crashed two nights ago. I found a 33 meg file from the boss containing last month’s presentation), I came upon something I mailed to myself, this link to various graphical representations of information spaces. In case anyone was interested.

Daring Fireball: On Joel on Software

Mon, 23 Sep 2002 12:18:21 GMT

Picking apart Spolsky[base ']s argument bit by bit, while fun, misses the forest for the trees. Windows software ported to the Mac almost always fails. Mac software ported to Windows very often succeeds. On the Mac, in any given software category, the best app usually wins. On Windows, in any given software category, Microsoft usually wins.

Thu, 19 Sep 2002 22:05:32 GMT

At the Whitney, Secrets of Digital Creativity Revealed in Miniatures.[New York Times: Arts] See the computer code that created the digital art. Mirapaul’s attitude denigrates the idea that computer code can be a creative expression, the same sort of attitude that is criminalizing DeCSS, certainly an elegant piece of code.

Mon, 16 Sep 2002 22:54:05 GMT

I could have sworn I posted this before: Teddy, a three-dimensional modeling tool. I don't remember its provenance: probably either the Squeak Smalltalk general mailing list or just surfing. Searching for those that link to it on Google, though, got me an interesting digital video site for teachers.

Mon, 16 Sep 2002 04:53:22 GMT

Can’t afford a Mac? Build one.