Despite a half-century of practice, a distressingly large portion of today's software is over budget, behind schedule, bloated, and buggy. To those who wonder why, and whether anything can be done about it, I have long recommended the book The Mythical Man-Month, by Frederick P. Brooks, Jr. http://www.amazon.com/exec/obidos/ASIN/0201835959/ This book has stayed continuously in print since 1975, and remained remarkably relevant. Now there is another book I would put beside it. A little more technical and less management-oriented, but equally thought-provoking. It is Software Fundamentals: Collected Papers by David L Parnas, Daniel M. Hoffman and David M. Weiss (eds.), Foreword by Jon Bentley: http://www.amazon.com/exec/obidos/ASIN/0201703696/ Parnas has been writing seminal and provocative papers about software and software development for more than 30 years, and this book collects more than 30 of them. It includes well-known classics such as "On the Criteria to Be Used in Decomposing Systems into Modules," "On a 'Buzzword': Hierarchical Structure," "On the Design and Development of Program Families," "Designing Software for Ease of Extension and Contraction," "A Rational Design Process: How and Why to Fake It," and "Software Engineering: An Unconsummated Marriage." It also has some lesser-known gems, such as "Who Taught Me About Software Engineering Research?", "Active Design Reviews: Principles and Practices," and "Software Aging." Browsing or reading this book, I think you'll be struck with how much of today's "conventional wisdom" about software was introduced (or championed very early) by Dave, and by how many of his good ideas have still not made their way into current practice. (Why?) Parnas isn't always right, but he's never dull. One of the most valuable things to do with this book is to pick something he says that you disagree with, and try to construct a convincing argument that he's wrong -- you'll probably find it harder than you expect, and you'll almost surely learn something valuable. Jim H. PS. Truth in advertising: I wrote introductions for two of the papers, but I don't get royalties.
The March 2001 issue of the *Communications of the ACM* contains an article by Edsger Dijkstra called "The End of Computing Science?" In it, he states "I would therefore like to posit that computing's central challenge 'How not to make a mess of it,' has *not* been met." As many of the RISKS entries have shown, application and other developers have certainly made a mess of things at times, often of Laurel and Hardy proportions ("That's another fine mess you've got us into."), and worse. If/when Software Engineering becomes a fully licensed profession, perhaps part of the code of ethics should be similar to the intent of part of the Hippocratic Oath, "First, do no harm". This is a paraphrase of the statement "The health and life of my patient will be my first consideration" which is from the World Medical Association's "Declaration of Geneva" of 1948. Or, as colleague Glen McCort once said in a meeting, "Don't do anything really stupid." Michael Cook [There is a big difference between Hippocrates and Hypocrites. In particularly, there are quite a few Hypocrites who claim they are "Software Engineers" but nonetheless write extremely riskful software. PGN]
I was in Chur in Switzerland last week and read the sad story of the lost train in the local newspaper. They were having trouble with a train that had to be diverted because of technical troubles along the line. Someone made a mistake while entering in the departure times in their tracking system. The system complained, something along the lines of: "You can't enter a departure time that has already passed", but someone pushed "do it anyway", and somehow managed to get the train sent off. They called, manually, each station along the (beautiful and scenic) route to Chur to let them know that the train was coming. No problem, except that someone forgot to tell the penultimate stationmaster. Since he did not know the train was coming, he dispatched the last little train of the evening off to the skiing resort Davos, and was packing up his things to go home when the train came into his station. Imagine his shock! There were still 5 passengers on the train that wanted to get home. Apparently it took quite a lot of discussion before everyone managed to get a taxi home, courtesy of the Swiss National Train Company. [Rhaetian Railway? See RISKS-21.44. PGN added in archive copy.] Just goes to show you: If people think they have entered in something correctly, no amount of error messages will convince them otherwise. Prof. Dr. Debora Weber-Wulff, FHTW Berlin, FB 4, Treskowallee 8, 10313 Berlin GERMANY +49-30-5019-2320 http://www.f4.fhtw-berlin.de/people/weberwu/ [Not quite Chur-noble, but perhaps Chur-lish. PGN)
The recording industry may be hoisted on its own petard if the Napster-like music swapping service called Aimster is successful in its legal strategy against the Recording Industry Association of America (RIAA). Unlike Napster, Aimster (which has no central servers to maintain and leaves users individually responsible for their actions) encrypts transmissions, and so there is no way for the RIAA or any other outside party to distinguish between files which are in compliance with copyright law and those that infringe on it. Of course, RIAA could simply decrypt the files -- but then it would be in violation of the Digital Millennium Copyright Act (DMCA), a law that it strongly supports, and that makes it a criminal offense to circumvent encryption protection of copyrighted material. (*The New Republic*, 21 May 2001; NewsScan Daily, 21 May 2001; http://www.tnr.com/cyberlaw/babbitt051101.html [NB: Correct English usage is: "hoist with one's own petard" (victimized or hurt by one's own scheme) (Webster via PGN)]
Newsgroup: alt.math.recreational WARNING: Do NOT calculate Pi in binary. It is conjectured that this number is normal, meaning that it contains ALL finite bit strings. If you compute it, you will be guilty of: * Copyright infringement (of all books, all short stories, all newspapers, all magazines, all web sites, all music, all movies, and all software, including the complete Windows source code) * Trademark infringement * Possession of child pornography * Espionage (unauthorized possession of top secret information) * Possession of DVD-cracking software * Possession of threats to the President * Possession of everyone's SSN, everyone's credit card numbers, everyone's PIN numbers, everyone's unlisted phone numbers, and everyone's passwords * Defaming Islam. Not technically illegal, but you'll have to go into hiding along with Salman Rushdie. * Defaming Scientology. Which IS illegal -- just ask Keith Henson. Also, your computer will contain all of the nastiest known computer viruses. In fact, all of the nastiest POSSIBLE computer viruses. Some of the files on my PC are intensely personal, and I for one don't want you snooping through a copy of them. You might get away with computing just a few digits, but why risk it? There's no telling how far into Pi you can go without finding the secret documents about the JFK assassination, a photograph of your neighbor's six year old daughter doing the nasty with the family dog, or a complete copy of the not-yet-released Pearl Harbor movie. So just don't do it. The same warning applies to e, the square root of 2, Euler's constant, Phi, the cosine of any non-zero algebraic number, and the vast majority of all other real numbers. There's a reason why these numbers are always computed and shown in decimal, after all.
Although it is not directly computer relevant, this case is nonetheless noteworthy in RISKS, where April-Fools' spoofs and parodies are an old tradition. A U.S. appeals court in Atlanta today overturned a lower-court ruling that Margaret Mitchell's estate could block the publication of ``The Wind Done Gone'', an apparent parody of ``Gone With the Wind'' that is written from the point of view of black slaves. [Source: Karen Jacobs, Reuters, 25 May 2001, PGN-ed]
The Federal Bureau of Investigation has in the past ten days charged 88 individuals with Internet crimes, including wire and mail fraud and money laundering. A government prosecutor said: "Internet fraud - whether it's in the form of securities and other investment schemes, online auction and merchandising schemes, credit card fraud and identity theft - has become one of the fastest-growing and most pervasive forms of white-collar crime." (Bloomberg News/*The Washington Post*, 24 May 2001; NewsScan Daily, 24 May 2001; http://washingtonpost.com/wp-dyn/articles/A67744-2001May23.html)
When Richard Smith (Privacy Foundation's CTO) obtained his FBI file from Choicepoint in Georgia, he discovered that he had died in 1976, and had had aliases with Texas convicts known as Ricky or Rickie. This is apparently the kind of info that the FBI now depends on. In 1998, a Chicago woman with no criminal record was fired after Choicepoint info mistakenly indicated she was a shoplifter and convicted drug dealer. Choicepoint info was also involved in thousands of Floridians being mistakenly identified as felons and disenfranchised in the November 2000 election. Choicepoint blames that on a data aggregator, DBT. [Source: Julia Scheeres, What They (Don't) Know About You, 11 May 2001 http://www.wired.com/news/privacy/0,1848,43743,00.html; PGN-ed] [With regard to flagrant data mining of incorrect information, What's yours is mined. PGN]
According to an article in The Register, the Council of the European Union is considering implementing rules that call for storing all telecom traffic (all phone calls, all Net usage, every e-mail) and making this data accessible for at least seven years. This will be done in the name of "public safety and law enforcement," no doubt. http://www.theregister.co.uk/content/5/19003.html Technical considerations aside (the concept of server farms the size of France comes to mind), the whole thing is just a dreadful idea. Dave Weingart, Randstad North America firstname.lastname@example.org 1-516-682-1470
The Web site of the federally funded Computer Emergency Response Team (CERT) was clogged by a "denial of service" attack that lasted 30 hours this week. CERT, which is located at Carnegie Mellon University in Pittsburgh, has a mission of providing warnings about computer attacks and viruses. An official of the organization said: "We get attacked every day. This is just another attack. The lesson to be learned here is that no one is immune to these kinds of attacks. They cause operational problems, and it takes time to deal with them." [AP/*USA Today*, 24 May 2001; NewsScan Daily, 24 May 2001 http://www.usatoday.com/life/cyber/tech/2001-05-24-cert-hacked.htm]
(by way of David Kennedy) Courtesy of the FAA: The FAA has this neat airport traffic website: http://www.fly.faa.gov/flyFAA/index.html where you can check out conditions at any airport. Well, recently they added the option to get e-mail on airport conditions: http://www.fly.faa.gov/Notify_Signup/notify_signup.html with a warning to be careful not to select all airports as that would be a lot of mail. Now the way this works is you put in an e-mail address and a password. this is the password to make changes on the FAA's site. Then they ask you what airports and how many characters your e-mailer can handle. I have selected DTW and for days I will get no mail. This morning I have already gotten 3 messages about various delays due to different thunderstorms. SO if someone does not like someone else, they just set this system to mail bomb the other person's cell phone. Imagine how annoying it will be with a phone constantly going off and not knowing how to stop the mail. would most people figure out how to get this stopped? **I** have not contacted my cellular provider on how to stop SMS spam, so I doubt if there is much experience here. there will be before this year is done. Robert Moskowitz, Senior Technical Director email@example.com ICSA Labs, a division of the TruSecure Corporation (248) 968-9809
[From Woody's Office Watch (http://www.woodyswatch.com)] 4. IN OFFICE XP, THE LINK YOU TYPE AIN'T WHAT YOU GET Remember when I asked you to send me your rants about Office XP? Editor-in-Chief Peter Deegan has a great one: I didn't believe it when it first happened to me, but now Microsoft arrogantly and shamelessly confirms the bug. When you type a hyperlink in FrontPage 2002, Word 2002, Excel 2002, PowerPoint 2002, or Outlook 2002 (using Word as your email editor), the Office application will alter what you've typed, without notifying you or giving you an opportunity to undo the "correction." In fact, in most cases, you can't override the "correction" at all: you're stuck with FP, Word or Excel's version of what you typed. Tough luck Charlie. Try it yourself. In Office XP, choose Insert | Hyperlink then type in this fake hyperlink http://www.fred.com/trial//2345/ Hit enter, and the double slash is unceremoniously converted to a single slash. You aren't notified. You aren't given a chance to change it. In fact, with one exception, you can't even *override* Office's ham-handed mangling of your carefully constructed hyperlink. The exception: in FrontPage 2002 you can fix the link by going into HTML mode and overtyping - but there's no such option in Word, Excel PowerPoint, or Outlook. Even Microsoft can't suggest a workaround. It's even worse than you might imagine. The text appears in the document the way you typed it - that is, you'll see http://www.fred.com/trial//2345/ in your document. But the link itself - the part behind the scenes that controls where you go when you click on the text - is altered to http://www.fred.com/trial/2345/ without any notice. Don't believe me? Follow these instructions, then right-click on the hot text and pick Edit Hyperlink. Look in the Address box. See that? While a double slash is unusual, it is a valid hyperlink used in the real world, most commonly as a delimiter between parameters. Microsoft has no right to arbitrarily change a link I've typed, especially if there's no way to override the change. We put this problem to Microsoft's PR folks with a series of questions to help clarify the situation. Their response was among the most arrogant and obfuscatory we've seen in many years of dealing with the company - a dismissive response not designed to help or reassure prospective Office XP purchasers. In fact, it has only made a bad situation worse. Microsoft says it's not an issue at all! The change is done intentionally for (you gotta love this) "cleanliness and consistency." Oy. Apparently the accuracy of a hyperlink is secondary to it looking nice. Microsoft dismisses the double-slash change problem saying they "don't know of any servers which deal with a double slash in the path component any way other than to treat it as a single-slash". C'mon. Call 1-800-GET-A-CLUE guys. Double slashes are used all the time. More than that, it isn't Microsoft's job to decide whether the URLs I type are politically correct. Microsoft goes on to say "some older servers did not like to have the double-slashes in the path and had difficulties with double slashes." Well, OK, that may be true but there are plenty of other typing errors that can make a link break. Double-slashes may be a problem in some cases, but in others they are necessary. I really wanted to hURL when the 'Softies said, "we don't change the parameter data, only the path part of the URL." Good grief. This comes from a company that assumes everyone uses the Microsoft method of passing information through links. In the Microsoft world you pass data to a web page by adding a question mark to the end of the link then adding the variables. Incredibly, not everyone uses Microsoft servers, and there are other ways to pass information through a web link. One of the ways we've found includes having double-slashes. Microsoft Office XP now blocks those uses with no recourse. Even if you accept the logic that double-slashes in hyperlinks are non-existent or bad, that doesn't change the more general principal that the user is entitled to type in something and have it stick, unchanged. If Microsoft wants to make a change for "cleanliness and consistency" they should build in a warning to the user and a way to reverse the change. A Smart Tag would work nicely. But in this case neither of these basic design courtesies is honored. The company has gone too far in compulsory changes to the link with no warning to the user or any workaround to fix the Autocorrect. Adding injury to insult, there's no documentation on these changes in the help file. Microsoft has declined to provide details of any other compulsory changes made to hyperlinks in Office XP nor have they suggested any workaround for those affected, or some way to switch off this behavior. The Microsoft arrogance shows through: it's not a problem, so why bother fixing it? The fact that Microsoft has declined to detail what changes are arbitrarily made to links makes us even more concerned. Office XP users don't know what compulsory changes will be made to their links. Chances are they'll find out the way I did - the hard way. Jonathan Arnold firstname.lastname@example.org Senior Product Developer Integrated Delivery Systems http://www.smartdrops.com
Someone recently sent me a reference to a program called Weatherbug and asked me to evaluate it from the perspective of a network admin for a small company where some employees are using it. It's a Windows program that places a local temperature icon in your taskbar and then continuously monitors local weather data from the AWS Weathernet. If you click on the taskbar icon it displays a panel showing local weather data updated in near-real- time. The service and Weatherbug executable are free and the whole thing is supported by advertising that is displayed in the Weatherbug window. I was curious about the security implications so I downloaded and installed Weatherbug with the intention of monitoring the IP traffic it generates with a packet sniffer. The first thing that happens during install is you are asked if you want to also install two additional tools, "Gator" and "Offer Companion". Here's the blurb on the install dialog: By including Gator and its OfferCompanion Software with Weatherbug, we're making your computer smarter! Gator and OfferCompanion are among the web's most popular products. Gator fills in your passwords and online forms automatically - with no typing! And OfferCompanion delivers great offers to you based on web sites you visit! The checkbox indicating that you want to install these "products" is checked by default. Needless to say, I did NOT allow it to install them (but then how do I know whether it listened to me or not ;-). Gator is clearly dangerous. I assume it keeps a database of previously seen web forms and the data you entered previously, and then re-enters the same data the next time you visit the same page. Regular RISKS readers should be cringing visibly by now :-) Anyway, I started up Weatherbug and monitored its traffic: 1) During registration you are asked to provide quite a bit of personal info, including name, address, and income. Luckily (or I wouldn't have proceeded) all data is optional except for your Zip code, so it can locate weather stations nearby. The registration data is sent to a Weatherbug server in an HTTP GET request. 2) After you register, the software sends an HTTP POST to 18.104.22.168, which does not seem to have a reverse DNS entry. The POST data is: InstallType=Full+Install&GatorStatus=Opt-Out&BCheck= 3) It appears to do everything over HTTP, so it's totally "pull" based. It does not *appear* to open any persistent connections. Also it seems to issue only GET requests in normal operation. I didn't see any POSTs other than the one described above. Of course, it's quite possible to send any data as parameters in a GET, so the absence of POST shouldn't be taken as implying anything positive. 4) In addition to retrieving weather data from the location you configured (any of over 5000 AWS sites located mainly at schools), it downloads ad gifs from doubleclick.net. 5) During registration you are assigned a registration ID that is sent to the Weatherbug server at various times. I did not see any evidence that the registration ID is sent to sites other than Weatherbug (i.e. ad requests didn't include the registration ID) 6) Every time Weatherbug starts up, my Win2K machine issues a single NETLOGON request to the PDC with a blank username, which is rejected. I don't know enough about MS authentication protocols to know if Weatherbug is doing this or it's just a byproduct of how Windows works. 7) When the main window is hidden (to a taskbar icon), most IP traffic stops. I still checks the weather data about once a minute but does not appear to load ads. 8) If you uninstall and re-install Weatherbug you are not asked to register again. The uninstall does not delete registry keys, so in order to completely remove it you must manually edit the registry. I found no evidence that Weatherbug is "spyware", but then this was a very cursory examination. It does seem to limit its data capture to your direct interactions with its GUI, but the possibilities for abuse are so high that I would not personally use it on an ongoing basis. It include an automatic software update capability and there's no guarantee that future versions won't quietly slip in some "enhanced" data gathering techniques. When the capability is there, the temptation to use it has got to be tremendous. Beyond the obvious security risks I'm also concerned about Weatherbug's bandwidth usage. When the main window is open and updating both weather data and ads in real time, it consumes about 20 kilobits/second. If you're a small company depending on an ISDN, DSL or fractional T1 link, it doesn't take very many of these to adversely affect other users. I'm curious to know if anyone else has conducted a more thorough evaluation and analysis of Weatherbug. James Garrison, Athens Group, Inc., 5608 Parkcrest Dr, Austin, TX 78731 email@example.com 1-512-345-0600 x150 http://www.athensgroup.com
Software piracy grew in 2000 for the first time in more than five years, according to the Business Software Alliance, which estimates that 37% of all software programs used by businesses worldwide are illegal copies. The Asia-Pacific region -- where more than half of all software in use last year was stolen -- tops the list in terms of dollars (an estimated $4 billion) lost to piracy. Meanwhile, Eastern Europe has the highest piracy rate, with 63% of its software illegally copied in 2000. In the U.S., 24% of programs are pirated copies. Although progress is being made in some regions, BSA director of enforcement Bob Kruger takes little comfort. "That's kind of like saying that I'm having fewer heart attacks than I used to. But the damage that's being caused by piracy is still devastating. It can be counted in the thousands of jobs and billions of dollars lost." (AP 21 May 2001; NewsScan Daily, 21 May 2001; http://news.excite.com/news/ap/010521/07/software-piracy ]
Please report problems with the web pages to the maintainer