Monthly Archives: August 2012

Aggressive Assembly Language Programming in SPITBOL

The Preface to “Macro SPITBOL: The High-Performance SNOBOL4 Language,” by Mark B. Emmer and Edward K. Quillen, includes the following:

Robert Dewar created Macro SPITBOL in the mid 1970’s, and showed the world that a high-performance version of SNOBOL4 was possible. His Macro SPITBOL compiler forms the kernel of this implementation. As an example of “aggressive” assembly-language-programming, it remains a programs tour de force.

I was reminded of this when I came upon a comment by Mark in the source for Macro SPITBOL:

note: as an experiment, we tried aligning all ent and prc’s on a dword boundary, in an attempt to minimize instruction stall time on the first opcode of a procedure. the resultant exe file waslarger by 592 bytes, and actually ran 0.1% slower. we will continue to use odd alignment of block entry routines so that the .cepp conditional may be used.

592 bytes? 0.1% difference in performance?

Can you think of another programmer who pays this level of attention to performance?

Here is another:


* move chars from xl (esi) to xr (edi), count in wa (ecx)
*
* the following sequence "old method" is shorter than the "new method"
* shown below, but is much slower because of the conditional jumps that
* cause the instruction cache to be flushed for 3 out of 4 count values.
*
* old method:
* shr ecx,1
* jnc tmp1
* movsb ; move odd byte
* tmp1 shr ecx,1
* jnc tmp2
* movsw ; move odd word
* tmp2 rep movsd ; move string as double words
*
* genop('shr','ecx','1')
* genop('jnc','short ' (t1 = genlab()))
* genop('movsb')
* genopl(t1 label.delim,'shr','ecx','1')
* genop('jnc','short ' (t1 = genlab()))
* genop('movsw')
* genopl(t1 label.delim,'rep','movsd') :(opdone)
*
* new method:
* shrd eax,ecx,1 ; preserve ecx[0] in eax[31]
* shr ecx,2 ; preserve ecx[1] in cy, divide by 4
* rep movsd ; move dwords, leaves ecx=0
* adc ecx,ecx ; copy cy to ecx[0]
* rep movsw ; copy 1 or 0 words, leaves ecx=0
* shld ecx,eax,1 ; copy eax[31] to ecx[0]
* rep movsb ; copy 1 or 0 bytes

elel

Some knowledge — indeed more than I have — of programming for Intel’s x86 architecture is needed to fully appreciate the above, but the attention to detail is obvious.

Dewar’s MACRO SPITBOL was preceded by SPITBOL/360, written in IBM360 Assembly Language in the late 1960’s by Dewar and Ken Belcher. [1]

It must be the most efficient 360 assembly language program ever written. Robert once remarked that it was the only program known to him where the authors kept a copy of the instruction timing/cycle details for each model of the 360 in front of them as they wrote the code, and they frequently consulted these details to decide on best code sequence.

SPITBOL/360 demonstrated an astounding level of performance. SPITBOL/360 routinely compiled hundreds of thousands of lines of source code per second.

I once heard, perhaps from Robert himself, of one of the great tricks in SPITBOL/360.

SNOBOL has an integer variable &STLIMIT. A counter is initially set to this value. Whenever a new statement is executed, the counter is decremented, and if the value reaches zero then an exception is raised. This is used to avoid runaway loops and other code that would result in execution going on forever.

The obvious method is to use an integer counter, decrement it when a new statement is executed, and raise the exception when the value zero is reached.

Dewar and Belcher’s device was to maintain the counter as a floating-point (real) variable. It was set by computing a “magic” value and initializing the counter to that value. When a new statement was executed, the value 1.0 was added to the floating point counter. The magic was in that a floating-point exception, due to overflow of the value, would be raised at the appropriate time.

Andy Koenig — a fellow contributor to Macro SPITBOL — made mention of this device in his column Some Programs Are Poorly Designed On Purpose. The post includes several other examples from SPITBOL of what Andy calls “nefarious purposes” in the use of floating point arithmetic.

Another device, one of much greater generality, was the use of indirect threaded code (ITC), described in Dewar’s paper Indirect Threaded Code.

Most interpreters use a “switch on opcode” statement when a new statement is to be execute. However, this involves testing the value, and doing a switch instruction to the appropriate code, which tends to slow things down. Indeed this is one of the main costs of writing an interpreter instead of generating actual code for the target machine. After all, the hardware is designed to extract opcodes and take the appropriate execution as fast as it is possible.

When using ITC, the transition from one abstract instruction to the next is made not by a switch, but by doing an “indirect branch” to the address contained in the the code block for the next statement.

By the way, this is the reason you cannot write code in C to execute the Minimal assembly language level code in which SPITBOL is written. C has no such contruct (what is needed is something like goto *p).

By the way, Dewar is equally aggressive when it comes to documenting his code. See for example the source code for SPITBOL in file spitbol.min available at Github Hardbol SPITBOL. I consider it a tour de force in program documentation.

For example, while working together on the port of Macro SPITBOL to the IBM PC in the early 80’s, I remember particularly an instance where Dewar saved a file, and then immediately went back to edit the source when he noticed there was an extraneous space..

Notes:

1. SPITBOL/360 is available in open source form under the GPL license. See SPITBOL 360.

Dewar and Belcher applied their aggressive programing techniques in the early 80’s by writing a COBOL compiler in COBOL! The wikipedia entry for Robert Dewar includes:

In the 1970s Dewar was a principal author of the Realia COBOL compiler, widely used in commercial environments to this day (marketed by Computer Associates).

Written in COBOL, a remarkable feat in itself, I learned later from colleagues at IBM that Realia COBOL generated code that was better than that produced by IBM’s product COBOL compiler

How to access, read, write, or mount a USB flash drive in Ubuntu Linux

My most viewed blog post is about how to access a flash drive on Ubuntu Linux. Written five years ago, it is my most-viewed post ever. For example, it now accounts for over a third of my views each day. [1]

This suggests that lots of folks are still having problems using flash drives on Ubuntu, and possibly other Linux distributions as well

The post is a bit dated. A lot has changed in the last five years, so here’s an update on how I currently access flash drives on my Ubuntu 12.04 desktop.

When you first put a flash drive into the computer all you will see is the light on the flash drive come on, if it has one.

If you do

$ ls /media

you won’t see any sign of the drive.

So what you do next is to open File Manager. It’s right there near the top of the main menu. It should show the flash drive. All the drives I have come with a built-in id, and File Manager will display it near the top of its page. Look for text that begins with /media/.

For example, I just plugged in a drive while writing this. It has the id 054A-FDA0.

Leave File Manager running, while you access the drive

If you again do

$ ls /media

you should see the drive. For example, I just did and found the file /media/054A-FDA0.

You can now do things like

$ cp /media/054A-FDA0/work.tar .

and so on.

It is good form to close the drive when you are done using it. This is not really needed if you are just reading the drive, but is important if you have written data to it. You want to know all the data has been written.

There are two ways to do what in Linux is called “unmounting” the flash drive.

One is

umount /media/054A-FDA0

Note the command is “umount” and not “unmount”. This is one case where Unix terseness, in this case saving a single letter, did more harm than good.

Another is to go back to File Manager, find the symbol for the drive in the left column, and right click your mouse on it. You will get a list of options, one of which is Eject Removable Medium. Then pick that, and remove the drive

Either will work. You make the call.

By the way, whenever you are typing a command that includes the flash drive id, as soon as you have typed the first letter or so, you should hit the Tab key. The shell will then complete the rest of the name for you.

(This is true whenever you are entering a file name. It’s one of the very nice features of Linux: thoughtful people have worked hard to make your work easier.)

Notes:

1 The second most-viewed post is A Brief History of Operating Systems, based on a couple of days of writing done while I was at IBM. Steve Mills, then and now head of IBM’s Software Group, felt that his salesmen didn’t know enough about this topic, and asked that someone put something together. The request landed on my desk, and not having anyone else I could kick it down the road to, I had to do it myself. [2]

2. One of the sagest observations I have ever heard came from Ralph Griswold, in a conversation almost forty years ago:

I wish I had done it myself. In the long run I always do.

Linux: The Last OS Standing

One of my favorite films is Akira Kurosawa’s Yojimbo. It stars Toshiru Mifune, who plays a lone samurai who comes into a town where two factions are feuding, and then skillfully plays one against another until he is the last man standing, so it’s no surprise that the film was (poorly) remade as Last Man Standing, starring Bruce Willis, in one of his worst films (all gore, no subtlety, which was the point of Yojimbo.)

I was reminded of this while doing a cleanup/refactoring of the code for Macro SPITBOL.

There has been almost no work on SPITBOL, save the port to Linux, in the last decade. Most of the code has been stable since the mid 80’s, save for ports to different architectures.

SPITBOL has a run-time interface written in C. While perusing this code I noticed the code had conditional text for various operating systems: AIX, BSD, LINUX, SOLARIS, and WINNT. Not Windows, but WINNT, a sign of the age of the code.

I then decided to clean up the code. There was no need for SOLARIS. It’s dead, though its new owner Oracle has not yet addmitted it.

AIX is also an outlier OS now, used only in large servers, so that went next.

BSD, UNIX and LINUX are all variants of UNIX.

That left WINNT. I then realized that, while much of the WINNT code would probably still work, I had absolutely no intereset in supporting it. That’s because Microsoft has not made any meaningful innovation, at least when it comes to Windows, in the last decade or so.

One thing (is it the only thing?) Microsoft does well is to support previous versions of its various OS’s, going all the way back to DOS. For example, the Windows/WINNT version of SPITBOL, last touched over a decade ago, still works, and so is in my view sufficient for Windows users.

I thus edited the source so there is *no* conditional text to distinguish operating systems, because there is only *one* OS of interest. It runs everywhere, and is where operating system innovation can still be found.

So I deleted *all* the conditional text, and changed the source to just use “Unix” instead of “Linux” in the version/achitecture description.

Needless to say doing that edit was a lot of fun. Bye bye, SOLARIS. Bye, bye AIX. Bye, bye Windows. Just Unix is enough for me.

The remaining issue is what hardware to support.

I see only two architectures of interest: X86 and ARM. SPITBOL supports X86 now, and I will port the system to ARM within a few months.

This is one result of working on legacy code. It gives you a sense of what was once important, but no longer is.

It also gives you greater appreciation for the survivors, and all the hard work behind them.

On Apple vs. Samsung: A few billion here, a few billion there, a few billion where? Austin!

The technosphere has been abuzz last few days about the outcome of the Apple vs. Samsung patent litigation. For example, see the following from the NY Times: Jury Awards $1 Billion to Apple in Samsung Patent Case and Apple-Samsung Case Shows Smartphone as Legal Magnet.

See also Steven Vanghan Nichols’s post in ZDNET, The verdict is in: Samsung vs. Apple. More articles from ZDNET on this affair can be found at Apple v. Samsung: A battle over billions.

I agree with Mr. Nichols on the meaning of all this: Apple vs. Samsung verdict: It doesn’t matter. I completely agree with the view expressed at the end of his article (emphasis added):

In the end, all of us, anyone who buys any technology anywhere, will pay higher prices. And innovation, what the patent system was designed to encourage, will only suffer until this utterly messed up system is totally revised.

Important news can be found in the recent articles about Samsung’s planned investment in a plant in Austin, Texas: Apple is expected to buy 1 of every 10 chips sold worldwide this year, Samsung to invest $4 billion in Texas iPhone, iPad chip plant , and Samsung invests $4 billion in Apple chip plant

.

Let’s get this straight:

In the same week that Apple wins a verdict against Samsung for over one billion dollars, Samsung announces it is investing four billion dollars in a plant that makes chips for Apple!

Why is this confusing news not so confusing? It turns out that Samsung is Apple’s sole supplier for the key chips used in its products: the Apple A4 and the Apple A5.

A verdict of one billion dollars against Samsung, and an investment by Samsung of four billion dollars in a new plant to make chips for Apple, all in the same week?

To paraphrase the late Sen. Everett Dirksen (R) if Illinois: [1]

A billion here, a billion there, pretty soon, you’re talking real money.

While a billion bucks is oodles of boodle to most folks, my guess is that in this case the one billion bucks that Samsung will have to pay to Apple, if the amount of the award is upheld, is just small change — “chump change” if you will — to Samsung, since Samsung expects to make well over four billion dollars by continuing to supply chips to the company that sued it.

While some may argue this is a “victory” for Apple, I see it as a major strategic blunder.

Though Apple and Samsung will continue doing business, to me the major news from this affair is that:

We now know that Apple is willing to sue its suppliers, something those suppliers, and Apple customers, will keep in mind going forward.

Apple should keep in mind what happened in another lawsuit in which a company sued a customer, SCO versus Daimler-Chrysler — bankruptcy.

I agree with most folks that this wasn’t about Apple vs. Samsung, but was just the latest in the battle that really matters: Apple vs. Google.

Apple uses ARM-based chips in its products. Google Android also uses ARM-based chips. So this “arms race” between Apple and Google is also about “ARM.” [2]

The iPhone and Android each use ARM-based chips: Apple and Google use the same hardware technology, so the only differentiator left is software. There are two ways to fight on the software front. First, by being more innovative or more efficient. Lacking that, all that’s left is patent litigation.

Everyone knows this. For example, that’s why Google bought Motorola. Not to acquire the technology, but to acquire Motorola’s patent portfolio.

Indeed, the next battle in this ongoing war has just begun: Man bites dog: Google sues Apple for patent infringement.

While these suits are a risky business, they are really about RISC, Reduced Instrution Set Computing, Reduced instruction set computing, the technology underlying ARM-based chips.

But that’s a post for another day, on Intel’s greatest strategic fear: that ARM will drive Intel’s business up, up, and away, into “The Cloud”, as they will be unable to compete with the key feature of RISC technology: low power consumption.

Notes:

[1] Sen. Dirksen didn’t actually this, but everyone thinks he did. See <a href="http://en.wikiquote.org/wiki/Everett_Dirksen"Wikiquote: Everett Dirksen.

[2] ARM is a company that specifies machine architecture and licenses implemetations of the architecture. It doesn’t make the chips, which is why I say “ARM-based” chips, not “ARM’s chips.”

On Programming and Marketing

Yesterday Redmonk’s James Governor (@monkchips) sent the following tweet:

Developer as Marketer – see Twilio http://monk.ly/RG5s9Q yo @dN0t nice work! via @josevalles49

Curious, I clicked on the link.It was to a new post by James: Developer as Marketer – see Twilio. The post contains several very useful links, but the one that most caught my fancy was a post by Rob Spectre,What a Hacker Learns After a Year in Marketing. It’s about Rob’s work over the last year working in marketing, not programming.

Rob’s article has four main points:

  • This S**t is Hard
  • Data wins arguments
  • Calendar Management is a Skill
  • You Can Learn to Schmooze

During my years at IBM I was involved in two projects — Jikes and Cloudscape — in which marketing played a key role, and so I’d like to share some of the lessons learned.

Jikes

I have written extensively about the Jikes experience in my Jikes Archives.

The first marketing challenge for Jikes was to convince IBM Research management to let us release Jikes for Linux in binary form. The key issue here was possible contamination issues that might result since Jikes had to link against libraries licensed under the GPL; so I spent a week reading the GPL, and various writings about it, to make the case that this was not an issue.

Those arguments were accepted, and Jikes for LInux in binary form was released in mid-July 1998. This resulted in a flood of downloads, since there was great demand for more support for Java on Linux.

Much of the reason for that growth related to Rob’s point: Data Wins Arguments My partner Philippe put it this way when it came to Jikes:
Speed Kills. From the early days in mid-1996, about a month after we were first able to compile “Hello World” in Java, when we first realized the Jikes showed some promise, Philippe said we should emphasize speed. He had known that selling Jikes just as a compliant compiler would not impress Research management, so he *always* worked to make Jikes blazingly fast.

When Jikes first appeared it was from 10 to 50 times faster than Javac, the standard Java compiler from Sun (the greatest speedup was on IBM’s AIX.) Jikes’s speed was our killer feature. No one else came close then, or has ever since.

The greatest marketing challenge was to convince to release the source code for Jikes. I solicited comments from the open community. One in particular, from Bjorn Ekwald, made the most convincing case. A key point in his argument was that there was no money to be made in tools such as Jikes; the money came from the applications you would be able to develop using those tools.

My key marketing point to management was that for IBM to be accepted as an full participant in the open-source community, then IBM would have to open-source code of significant value, using an accepted open-source license, even if meant we would have to right our own, and we did. Also, IBM would have to run the project for at least a year from an URL ending containing “ibm.com” and show by example that we would accept contibutions from third parties, and govern the project under accepted rules, with the goal of having a core team with control over the project that would have the majority of its members from outside IBM.

From the earliest days when Jikes for Linux became an issue, I reached out to members of the open-source community. I knew I was ignorant in this area, and so I would need their help to help market Jikes to the world. For example, whenever I saw an email of comment from someone about what they were doing, and I could see that they would think, or plan, differently if they knew Jikes was going to be open-sourced, then I would tell them privately about it, just asking that they not tell others since IBM marketing had planned a big blitz for the Jikes announce in early December, 1998,

My favorite memory of the revelations was a call I placed to a number I knew was somewhere in Europe. When I reached the developer and told him what was afoot and then asked for him not to talk, he then said, “I’m in a bar in Barcelona. Who could I tell?”

No member of the open-source community *ever* betrayed the trust I placed in them:


Not once if my years at IBM dealing with non-IBMers about open-source issues did anyone ever betray my trust, not once. And there were occasions when I revealed information of some sensitivity, in some cases knowing that if what I had revealed became public knowledge, then my days at IBM would soon end.

Not once!

Jikes was certainly a great success, if only for myself and Philippe. Indeed, my main personal motivation for getting Jikes out to the world was so the world could see what an extraordinary feat of programming Philippe had accomplished. (While I wrote some of the code, he wrote all the most difficult parts.)

We received a major Research award for our Jikes work in mid-2000. I think it was largely due to a letter from a senior IBM exec, Robert LeBlanc, who said in the letter he so kindly wrote that “Jikes was featured in *every* IBM presentation, talk or meeting about IBM and open-source during all of 1999.” [1]

However, the compliment I most value came from a woman whose name I can’t recall at the moment. She worked in the IBM Research marketing and press relations department. She called me a few months after Jikes was released as open-source to say that she was leaving for a new job, but wanted to let me know that when she first learned what I was trying to accomplish, she thought that success was extremely unlikely, and appreciated what we had been able to accomplish.

Cloudscape

I first learned that IBM would be realeasing the source for Cloudscape in the spring of 2004.

Cloudscape was a Java program that was, as one IBM marketing person later aptly named it, “a database in a JAR.” It let the programmer retrieve a data set and then issue SQL queries against it from within a browser. I first learned about in at a presentation in 1999 when Cloudscape was a private company. It was later acquired by Sybase, and then came into IBM’s hands when IBM bought Sybase. It was valued at about about $70m, and was the most widely-used component within IBM’s Websphere portfolio.

I soon learned that Cloudscape was released as a marketing move. Steve Mills, head then and now of IBM’s Software Group (SWG), had decided that IBM should emphasize the SMB — small-medium-business — market in 2004. The Cloudscape folks had come to Steve asking permission to release Cloudscape in binary form, and Steve had replied that in the SMB space there were no CIO’s, just developers, and that in order to win over those folks they would have to release the code.

By the way, the most prescient remark about marketing and software I ever heard Steve say was: Code Talks!

The day after Steve’s decision I was approached by the Cloudscape team in my role as a member of the Open Source Steering Committee (OSSC), the small team then responsible for managing all of IBM’s open-source activities, worldwide and across all divisions. I was the only non-lawyer on the team based in Somers, HQ of SWG.

Their question was basically, “Steve says we have to release Cloudscape as open-source. What’s an open-source?”

Knowing the value of the code and the importance IBM attached to its release, I worked on little else until early August, when the code was released as a donation to Apache, where it was renamed Derby by Apache’s Ken Coar.

Soon after starting work on Cscape, I had the great good fortune to meet Tami Cannizzano, then and now in IBM marketing. (She is now a Director of Marketing.) She was unfamiliar with open-source, and I remember quite well one our first meetings, in the lobby of SWG HQ, when I first talked to her about open-source, Cloudscape, and how IBM should go about getting it out into the wile.

My first, and always key, point was that for IBM to succeed it was necessary to give the code to an established group, since the then planned approach — starting our own web site and having the project run by IBMer’s would be doomed to failure — since it would be so hard to enlist developers.

Apache was the obvious choice.

Also, in mid-May, I learned from Ken Coar’s management that Ken’s days at IBM were numbered. I had worked with Ken before during the Jikes days. IBM hired Ken for his open-source expertise back in 1998 or so. However, the value of that background had lessened over time as IBM had become more experienced in open-source issues.

The real problem was that, during an open meeting with senior management, Ken had the audicity to say that he much preferred Apache’s development process to IBM’s own development process. (As an experienced hand later informed me, mastering IBM’s process took years, and was not a task for the faint of heart.)

So I had a second, but much more important to me personally, marketing task: Saving Ken’s Job.

By mid-June I was able to introduce Ken to the team. He was asked to write his thoughts on what we should do, and responded with a brilliant note about the marketing challenges ahead and how we should face them. This was enough to get him a seat and the table, and he did a great job. (He also was able to keep his job. This was my greatest personal satisfaction from my years at IBM.)

In large corporations such as IBM, a *very* big deal is “The Announce,” the event at which something major is announced or released. The importance of this, at least to senior exec’s, cannot be over-emphasized.

As a marketer, this can used to your advantage. For example, the drafting of the Jikes license seemed an endless task. I stopped by the office of the attorney responsible for the license on Friday morning, three days before the planned announce. He said he still wanted to make some changes. I said, “No problem. Just don’t forget to tell the senior execs that they may have to postpone the release to be announced Monday at 1PM.”

He then said he was ok with the license in its current form.

You have to recognize marketing opportunities when they come your way, even though at first glance they may not seem like a marketing issue.

For example, sometime in May I was approached by folks from Zend. They said they were establishing a professional development program in PHP, and wanted the loan of some IBM machines that they would use at conferences to demonstrate their online training, and they they would be happy to publizice IBM’s support in this effort.

I soon learnd that it was *very* difficult to arrange loans of this sort. Fortunately, Tami was by then on board, and when I explained the importance of Zend and PHP in the open-source community, she was able to secure the machines. I soon got permission, and sent them on their way.

Later, in mid-July, I learned that the then CEO of Zend would be in New York, and we arranged to meet in the city for breakfast. There I informed him about the upcoming Cloudscape release, and asked for his help, which resulted in Zend folks being on stage when Cloudscape was announced.

One of my most vivid memories from Cloudscape was a long phone call, about an hour or so, phone call I had with a senior member of Apache.

Key to The Announce is keeping it a secret until the Great Day arrives. As part of this IBM had asked Apache folks to keep the lid on. However, this senior poohbah questioned this approach once he learned about it, as he felt it violated Apache’s open process.

I thus called him up, and spent that hour or so on the phone, in which I explained the importance IBM attached to keeping the lid on. In order to establish my credibility, and to emphasize the importance IBM attached to this release, I shared a good deal of information about IBM and open-source, knowing full well that he knew that if he later decided IBM was not acting in good faith, then he could certainly cost me my job by revealing just a fraction of what I was telling him.

It was during Cloudscape that I had the great good fortune to work with Jon Prial and Sandy Carter, true masters of marketing.

Cloudscape was a success. The most important recognition from IBM came in the form of a “Star Award,” which was a major award in SWG for sales and marketing. Since Cloudscape came about as part of a marketing effort, Ken and I thus became one of the few programmers to have ever won the award, which consisted of an over-the-top long luxurious weekend in Vancouver.

The recognition I appreciated most came from a remark from my manager. He later said that one of the limitations on working on strategic plays of this sort was that one could not expect recognition, as the trick was — and I’m paraphrasing here — to do the work behind the scenes so that when all was revealed, it seemed so obvious, though at the start no one had known how to bring it off.

He said I was the reason IBM donated Cloudscape to Apache.

Notes:

[1] I first met Robert at a meeting in Toronto in the early 90’s. He was then head of a group of 60 or so folks working in compilers, and our team from Research had come to Toronto to share some new work we thought might be of interest to them.

We had a good morning of meetings, then a nice lunch, and met with Robert about 2PM. He must have heard back from his folks by then, so the meeting was brief. After a few minutes, he looked at his watch, and said that if we left now, we would be able to make our plane by 5PM. Knowing that the airport was but a short drive away, and so there was no reason to leave so soon, we knew we had been dismissed.

For years afterward, whenever we felt management was not appreciative of our work, we would say to each other, “If we leave now, we can get the plane by 5PM.”

I was told several times that Robert had accumulated a heap of tickets driving his Porsche down route 684 from SWG HQ in Somers to Corporate HQ in Armonk, and that he held the land speed record for shortest travel time between the two sites.

The Republican Dilemma: Which is worse? Endorsing rape, or raising taxes?

The lead story in today’s New York Times is about Missouri Representative Todd Akin (R), G.O.P. Trying to Oust Akin From Race for Rape Remarks. The reporters are Jonathan Wesiman and John Eligon.

The orgigins of the the affair are as follows:

Amid an uproar over provocative comments on rape and abortion that Mr. Akin made in an interview broadcast on Sunday, the National Republican Senatorial Committee declared that it would withdraw financial and organizational support for Mr. Akin, including $5 million in advertising already reserved for the fall. In the interview, Mr. Akin said victims of “legitimate rape” rarely got pregnant.

Given that his remarks were so far out of the bounds of reason and dignity, I found the “uproar” to be little more than a tempest in a teapot. For example,

At the same time, Republican candidates like Mitt Romney and Senator Scott P. Brown of Massachusetts either called for Mr. Akin to step aside or strongly indicated that he should. In a radio interview, the conservative host Sean Hannity pleaded with Mr. Akin to drop out. “Sometimes an election is bigger than one person,” he said.

(My own guess is that while Mr. Hannity thinks some elections are bigger than one person, he knows in his heart of hearts that his own importance is bigger than that of any electoral outcome.)

Republican interest groups expressed outrage. Some even said they would withdraw funding for his campaing. In my view, his remarks are so far over the line that they should instead have said they were going to double down on the election, committing to spending twice as much money as they had planned, all of it to be in support of his opponent, the incumbent Senator from Missouri Claire McCaskill (D).

The tepid nature of the response let me to wonder if there were another remark that he could have made that would brought far greater opprobrium and calumny on him down on his empty head, and now shattered reputation.

Just a few seconds led to the observation I’m about to share. I expect that many of you who are reading this have already thought of it.

Suppose that, instead of endorsing rape, Rep. Akin had said, “I have had a Come-To-Jesus moment. I now realize that the only way to make true progress in restoring our economy and making our nation better prepared to face the other great challenges facing us, is to work with the Democrats so we can pursue a balanced path of spending cuts combined with suitable tax increases.”

TAX INCREASES! Imagine the pillorying that would have occurred had me made that outrageous statement. He would have been shunned, replacing the “A” for adultery on Hester Prynn’s garment with “T” for Taxcutter on his.

Since Mr. Akin did win the Republican primary to run against Senator McCaskill, the Republican estabishment, such as it is, cannot force him out of the race.

What could be done however, would be to tell Rep. Akin that if he persists in his race and manages to win the election, then he will indeed be shunned. He will not be welcome at the White House if Mr. Romney wins. He will be shunned by the House leadership.

For example, the Senatee Majority Leader — assuming the Republicans gain control — could say he would be given only the most meaningless committee assignments.

But wait. He has already been given one of the most meaningless committee assignments, at least in the view of the current House Republican leadership. As Andy Borowitz noted in a recent tweet:

Was still trying to wrap my brain around Todd Akin being elected to the House 6 times & now I learn he’s a member of the Science Committee.

I’m now an authority on operating systems

Since I resumed blogging after a three-year absence, I’ve noted with some surprise that my most read post each and every day has been A Brief History of Operating Systems, written several years ago.

A quick vist to Google and a search on “history of operating systems” showed I am #3, almost *the* go-to guy in the world when it comes to the history of operating systems.

WordPress’s statistics suggest that most viewers of this post come from the Far East, notably India and the Phillipines. I expect I’ve been cited in many papers from students on the far side of the world who are taking a course on operating systems. I’m not quite sure what to make of this, though I do think that my use of the word “brief” in the title suggested an easy way out to students who use Google as their primary reference source.

This is yet another example of my being one of the Google’s most cited sources on “authoritative opinions,” as described in my 2007 post An Authoritative Opinion on Libraries and Authoritative Opinions.

The point made then is still true today: One can indeed game Google, sometimes in very surprising ways.

It doesn’t mean a damn, but is sure as hell is lots of fun, at least to me.

I wonder if I can finagle a world tour with a lecture series on operating systems based on a request to provide some guidance to IBM sales folks that I spent a couple of days putting together.

Maybe I should write about bringing Peace to the Mideast, or — even better — offer my services to interested parties on my authoritative opinion on who is best qualified to be our next President…

L’Affair Zakaria: Serge Lang on Fareed Zakaria

Today’s New York Times Business Section has two articles that make mention of Fareed Zakaria. A Media Personality, Suffering a Blow to His Image, Ponders a Lesson by Christine Haugh gives a good summary of the recent charges of plaigarism that led to Zakaria’s temporary suspension from Time magazine, and says in part:

But Mr. Zakaria’s career suffered an abrupt setback recently after bloggers discovered that his column of Aug. 20 for Time magazine had passages lifted almost entirely from an article by the historian Jill Lepore that appeared in The New Yorker in April.

Mr. Zakaria quickly apologized. But within minutes, Time had suspended him for a month and CNN, which had posted parts of the column on its Web site, removed the article and suspended him until further notice. Both began investigations of his work, as did The Post.

On Thursday afternoon, Time and CNN said they had completed their reviews, found no evidence of plagiarism and restored Mr. Zakaria to his demanding schedule. Just as quickly as his employers had questioned his credibility, they rallied around him.

“He’s one of the premier global intellectuals,” said Richard Stengel, managing editor of Time. “He will recover.”

Mr. Zakaria is also the subject of David Carr’s column Journalists Dancing on the Edge of Truth. It says in part:

Ten days ago Mr. Zakaria, who has a show on CNN and columns in Time and The Washington Post, acknowledged plagiarizing content for a column in Time. He apologized, was suspended, and Time and CNN investigated whether there was a deeper problem and decided there was not. He was reinstated on Thursday. End of story.

Mr. Carr later writes (emphasis added):

The now ancient routes to credibility at small magazines and newspapers — toiling in menial jobs while learning the business — have been wiped out, replaced by an algorithm of social media heat and blog traction. Every reporter who came up in legacy media can tell you about a come-to-Jesus moment, when an editor put them up against a wall and tattooed a message deep into their skull: show respect for the fundamentals of the craft, or you would soon not be part of it.

I once lost a job I dearly wanted because I had misspelled the name of the publisher of the publication I was about to go to work for. Not very smart, but I learned a brutal lesson that has stayed with me. Nobody ever did that for Mr. Lehrer, even after repeated questions were raised about his work.

It may not have made a difference: journalists are tasked as seekers of truth. Fabulists find the truth quotidian and boring, insufficient to convey them to the renown they seek.

But as rapidly as the Web can indict, it can also rehabilitate. Mr. Zakaria went from abject apology to justification to reinstatement in a matter of days, all in real-time bulletins. What was once a gallows, a place of professional execution, has become a kind of highly visible penalty box.

Reading the articles reminded me of a talk I once heard by the eminent Yale mathematician Serge Lang. Prof. Lang died in 2005. Were he alive today I know he would have written about this affair, because he mentioned Mr. Zakaria several times during his talk.

I was at the talk because my youngest daughter Jennifer is a Yale graduate (JE ’06). As with almost everything Yale does, the visiting days for parents were well-done and always a treat, as one had a chance to heard some of Yale’s best teachers in person.

One year, probably it was 2004, I noted that Serge Lang was speaking. I knew of his work, especially the clarity and vigor of his writing, and so went to the talk.

Lang was born in 1927. His family moved to California and he, like me, did his undergraduate work at Caltech. He went on to get his doctorate at Princeton under the great Emil Artin. He made many contributions, and among his awards was the Steele Prize for Mathematical Exposition, awarded by the American Mathematical Society. He was also a member of the Bourbaki Group. “Bourbaki” was the pseudonym used by a group of mathematicians who wrote a wonderful series of books on the mathematics of the 20th century.

He was also, as Wikipedia notes, an activist. He was an early opponent of the Vietnam War. He was also later, and much less presciently an Aids denier, questioning the theory that the HIV virus caused Aids.

In the 80’s he initiated another affair, described by Wikipedia as follows:

In 1986, Lang mounted what the New York Times described as a “one-man challenge” against the nomination of political scientist Samuel P. Huntington to the National Academy of Sciences. Lang described Huntington’s research, in particular his use of mathematical equations to demonstrate that South Africa was a “satisfied society”, as “pseudoscience”, arguing that it gave “the illusion of science without any of its substance.” Despite support for Huntington from the Academy’s social and behavioral scientists, Lang’s challenge was successful, and Huntington was twice rejected for Academy membership. Huntington’s supporters argued that Lang’s opposition was political rather than scientific in nature

I had gone to his talk expecting he would tackle some topic of mathematics, as I was curious to see how he would present it to a group of parents, most of whom presumably had little training in mathematics.

However, he chose another topic, that of the difference between the “soft sciences” such as psychology and sociology and the “hard sciences” such as physics and mathematics. His main point, as best I can recall, was that too often social scientists claimed a rigor and certainty in reporting their findings as “science”; which he felt was unjustified in that they had not used the scientific method, at least as he understood it.

He spoke at great length on the Huntington affair mentioned above. I expect most parents there felt that they had wandered into the wrong talk. I cut him a good deal of slack due his reputation as a world-class mathematician.

He also made mention of Zakaria, who worked with Huntington.

Lang wrote about this in lang_on_zacharia.pdf (It’s easy to find via search for “Serge Lang on Zakaria” though I couldn’t get the exact URL.)

A thoughtful post on Lang can be found at Serge Lang vs. Bob Somerby in a BLOWOUT!, which says in part:

Besides math, Serge Lang is famous for two things: claiming that the link between HIV and AIDS has not been established, and keeping Samuel Huntington out of the National Academy of the Sciences. Lang gave me a few hundred pages of documents about the conflict, as he did with anyone who was willing to listen to him. He claimed that Huntington’s papers were “utter nonsense.” His biggest objection was to a paper Huntington wrote that purported to demonstrate the link between a society’s frustration and instability. One of his indices classed South Africa as a “satisfied society.” Lang thought (rightly) that Huntington’s effort to quantify oppression and instability didn’t correspond to reality. Huntington’s defenders typically turned this into a straw-man argument. They said that Serge Lang objected to any attempt to turn things like frustration and instability into numbers, and that the argument was caused by a mathematician’s resentment of the “soft” sciences. (See Jared Diamond’s Soft sciences are often harder than hard sciences in Discovery.) Lang didn’t actually have any problem with using numbers to measure satisfaction; his objection was that Huntington’s index in fact measured nothing. Lang particularly hated Fareed Zakaria, now the editor of Newsweek, who wrote a letter saying that it was “a fact” that in the sixties, there were no “major riots, strikes, or disturbances” in South Africa. Lang had a file of New York Times articles on South Africa, all contradicting Zakaria.

See also Domenico Rosa’s Pseudo-Science and Junk Scholarship , which says:

For many years Serge Lang, a professor of mathematics at Yale and a member of the National Academy of Sciences, has been trying to expose the dangers posed by purveyors of pseudo-science and junk scholarship [these are his phrases].

It is unfortunate that he and others have not devoted some of their energies to speaking up about the flimflam being promoted by the “math reform” and “education research” leadership.

I met Lang once at the fall 1987 meeting of the Northeastern Section of the MAA, when he delivered an amazing address entitled: The effectof abuse of so-called “mathematics” in some social “sciences.”

In 1978, Ladd had exposed the defective Ladd-Lipset survey of the American Professoriate–a survey that he described as “invalid at best, dangerous at worst.”

In 1986-87, he blocked the election of Samuel P. Huntington, Eaton Professor of the Science of Government at Harvard, to the National Academy of Sciences. Lang exposed all sorts of mathematical nonsense in Huntington’s book “Political Order in Changing Societies.” This book was required reading in Political Science 111b, which was taught at Yale by professor N. Bradford Westerfield.

Subsequently, Lang became the victim of a scandalous counterattack. In a letter dated June 8, 1987, the then provost William D. Nordhaus made
a poorly-disguised attempt to muzzle Lang. On July 27, 1987, Fareed Zakaria, a Yale graduate and Huntington underling, tried to discredit
Lang in an article published in The New Republic (“The professor’s vendetta. Blood Lust In Academia”). This diatribe backfired and
provided additional evidence of Huntington’s and Zakaria’s ignorance of ninth-grade algebra and South African history.

On Nov. 13, 1987, the Yale Daily News published a distorted article concerning Lang’s positions (“Lang cites lack of academic standards”). When Lang submitted a response, the editors attempted to censor him, and he was forced to publish his article as a paid advertisement on Nov. 19 (“Academic, Journalistic, and Political Problems”).

(I apologize for including so much text from the writing of others about Lang, but I didn’t take notes during his talk, and so have included these words about Lang because they reflect the strength of his passion and the force of his arguments.)

  • Pages

  • August 2012
    M T W T F S S
    « Jul   Sep »
     12345
    6789101112
    13141516171819
    20212223242526
    2728293031  
  • RSS The Wayward Word Press

  • Recent Comments

    dave porter on On being the maintainer, sole…
    daveshields on On being the maintainer, sole…
    Paul Tallett on On being the maintainer, sole…
    mrrdev on On being the maintainer, sole…
    KT on On being the maintainer, sole…
  • Archives

  • Blog Stats

  • Top Posts

  • Top Rated

  • Recent Posts

  • Archives

  • Top Rated