Female Mathematicians: Emmy and Sophie Have Lots of Company

Yesterday I grumbled that for Ada Lovelace Day, Smithsonian Magazine  had trotted out the same tired list of five historic women mathematicians: Hypatia, Sophie Germain, Ada Lovelace herself,  Sofia Kovalevskaya, and Emmy Noether. Nothing wrong with any of these estimable women (except perhaps Hypatia, evidence for whose existence is a bit shaky). Emmy Noether, in particular, was one of the undisputed giants of 20th century mathematics, responsible for important advances in abstract algebra.

Mary Jo Foley at ZDnet challenged me to come up with some fresh names. So here are 10 more contemporary women mathematicians you should know:

Julia Robinson (1919-1985), a student of the great Polish mathematical logician Alfred Tarski, she did important work in computability theory and made major contributions to the solution of Hilbert’s Tenth Problem, which proved that no algorithm exists to find integer solutions to general Diophantine equations. She was the first female president of the American Mathematical Society.

Ingrid Daubechies (1954- ) professor of mathematics at Duke, is probably the world’s leading authority on wavelets, a technique of tremendous improtance in signal processing.

Kathleen Synge Morawetz, (1923- ) professor emerita and former director of the Courant Institute of Mathematics at New York University, is known for her work in the field of the fluid dynamics of transsonic flows. She was the second woman to serve as president of the AMS.

Irene Stegun (1919-2008) was author, with Milton Abramowitz, of the National Bureau of Standards A Handbook of Mathematical Functions. For five decades, and especially before the widespread availability of computers, the Handbook’s tables of function values were an invaluable tool for researchers in pure and applied  mathematics. Although Stegun is listed as the second author, Abramowitz dies early in the project and the work is mostly her responsibility.

Mina Rees (1902-1997) earned her PhD at the University of Chicago under Leonard Dickson and did research in abstract algebra, but is best known for her work as director of the mathematical department of the Office of Naval Research, where she played an important role in the early development of computers. She was the first woman president of the AAAS.

Olga Taussky-Todd (1906-1995) was a leader in the modern development of linear algebra. During World War II, she did important work at the National Physics Laboratory in the UK studying the vibrations of aircraft.

Mary Ellen Rudin (1924- ), professor emerita at the University of Wisconsin, is best known for her contributions to  point-set topology.

Jean Taylor (1944- ), professor emerita at Rutgers, is a leader of research into the geometry of minimal surfaces, a field important in exlainting the growth of crystals–and soap bubbles.

Karen Uhlenbeck (1942- ), professor at the University of Texas-Austin, has done research in diverse fields inlcluding the caluculus of variations, nonlinear partial differential equations, and gauge theroy. She is a former MacArthur Fellow.

Fan Chung (1949- ) professor of mathematics at the University of California-San Diego and a leading graph theorist.

Women account for between a quarter and a third of the math PhDs awarded in the U.S., a number that hasn’t budged much in recent years. Women still have trouble getting tenured positions in top departments. And a woman has yet to win a Fields Medal or an Abel Prize. But these will come.

Tech.pinions: A New Place to Read My Stuff

I have become a regular contributor to Tech.pinions, a new site dedicated to news and analysis of important trends in technology. At Tech.pinions, I’m joining the founders, veteran industry analysts (and old friends) Ben and Tim Bajarin of Creative Strategies.

Tech.pinions is not going to cover every new gadget, let alone every rumor about every new gadget. It doesn’t do live blogs of event. The focus is on analysis, backed by both reporting and deep knowledge of the industry.

Give it a look. I think you’ll like it.

BlackBerry and iOS: Why First Movers Often Lose

Research In Motion was was ahead of everyone to the cloud. In fact, BlackBerrys were cloud devices long before we knew enough to call it “the cloud.” But failure to capitalize on that first-mover advantage has left RIM struggling to catch up in a game that it, in many ways, started.

Since the first BlackBerry, the 850 in 1999, was basically a two-way pager, RIM understood from the beginning that its success depended entirely on back-end services. The BlackBerry was a platform in a way that no other phone approached (Palm, in its heyday, came closest.) For a long time, the RIM back end outclassed anything else in the field, especially when a BlackBerry device was connected to a Microsoft Exchange messaging system through a BlackBerry Enterprise Server. When you got a new BlackBerry, the IT department sent you an authentication code. You ran the enterprise Activation app on the device, entered the code, and within a few minutes all of your mail messages, contacts, calendar information, todo lists, and notes appeared on your BlackBerry through the magic of the cloud. Over time, RIM added additional services, such as the ability to deploy custom app data and corporate documents via BES.

What RIM failed to do was to bring the slick, it-just-works beauty of an enterprise BlackBerry setup to consumers. As a client for a standard internet mail account, a BlackBerry is somewhat worse than average. Real-time, over-the-air calendar and contact sync required kludgey third-party solutions. And, of course, application development lagged far behind the offerings for iPhone and Android. Its biggest attractions for consumers, especially the text-centric, were outstanding keyboards and the excellent BlackBerry Messenger messaging system. But other hardware has caught up with blackBerrys in text entry and surpassed them in all other regards.

Today, platform religion is standard in the smartphone/tablet world. But the cloud components of Apple’s iOS5 and the tight integration among iPhones, iPads, iPod Touches, and Macs should put it in a class by itself. In one sense, Apple is 10 years late to this party, but once again, it has proved that getting there best is a lot more important than getting there first.

 

 

Are You Ready for the New Internet? Find Out Wednesday

On June 8, the folks who run the internet’s infrastructure will go live for the first time with Internet Protocol Version 6, a major overhaul in the protocols and addressing scheme that make the net work. Most users won’t notice anything different, but it’s a crucial step on the long road to IPv6. For a deeper look at what’s going on, read my News@Cisco post, “World IPv6 Day Will Test the Readiness for Change.”

How Real a Threat Is “De-Anonymization”?

“De-anonymization” is an ugly word and a scary concept. The idea is that if enough anonymous information about an individual is collected, mining the data can create a profile of a unique person, which can then be linked back to other public information to attach an identity to the data. This is clearly possible in theory and  has been demonstrated in research. But how much of a problem is it in the real world? And is the threat great enough to justfy restrictions on the collection of “non-personally identifiable information”?

The  question is not academic. It plays an important part in a case now before the U.S. Supreme Court, Sorrell v. IMS Health Inc. The case involved a challenge to a Vermont law that prohibits the sale of information on prescriptions identifying the prescriber without the doctor’s consent. On its face, the case is about doctors and commercial free speech, not patients and privacy. But amici curiae briefs filed by the Electronic Privacy Information Center, the Electronic Frontier Foundation, and others argue that the real issue is patient privacy.

In Internet Filtering Evil?

I have often marveled at how good Google–and Bing and Amazon–are at tailoring search results to my tastes and desires. Sometimes the uncanny accuracy can be a tad disconcerting, but on the whole, the personalization makes search far more useful. Until recently, I had no idea I was being cheated.

But in an op-ed in The New York Times, Eli Pariser, board president of MoveOn.com argues that personalized filtering is not only bad for me, it actually threatens democracy. Pariser’s argument, set forth at greater length in his new book, The Filter Bubble: What the Internet Is Hiding from You, is that individuals and society as a whole are done a disservice when search tools show them what they actually want to see. “Democracy depends on the citizen’s ability to engage with multiple viewpoints,” he writes. “The internet limits such engagement when it offers up only information that reflects your already established point of view. While it’s sometimes convenient to see only what you want to see, it’s critical at other times that you see things that you don’t.”

This is errant nonsense. There’s a theory about that the proliferation of specialized, sometimes ideologically narrow, sources on the net coupled with search filtering is leading to an increasingly polarized society. Our civic life is distressingly polarized, but this trend long predates the rise of the web, and I have yet to see anything resembling rigorous research that supports the notion that narrowly ideological web sites (or broadcasts) are a cause rather than a symptom.

Apple Should Protect Developers Against Lodsys Suit

Independent developers of iPhone and iPad apps that use Apple’s in-app purchase systems are being threatened withy legal action by a company that claims to hold a patent covering the technique. Like all patent disputes, this one promises to be long, complicated, and expensive. But Apple, which owes much of the success of its iOS products to these developers, should do the right thing and promise to indemnify them.

The threats are coming from a company called Lodsys LLC. which bought a portfolio of patents from inventor Dan Abelow. Lodsys is what’s known in patent-speak as a “nonoperating entity,” or more disparagingly, a “patent troll”–a company which seeks to derive revenues from intellectual property it owns but doe snot use in the course of its own business. I have no opinion on the validity or applicability of Lodsys’s patents or its demand for a royal of 0.575% of all in-app purchase revenues. My concern is with the relationship between Apple and the developers.

It’s clear that the developers have acted in good faith using technology for purchases that Apple not only provides but demands that they use. There’s also some evidence that Apple might have known there was a potential problem; Lodsys says that Apple itself (along with Google and Microsoft) has acquired a license to use its technology in its own branded products.  But, says Lodsys, “The scope of their current licenses does NOT enable them to provide ‘pixie dust’ to bless another (3rd party) business applications.”

Some providers of in-app purhcases, such as Time-Warner and Conde Nast, could certainly fend for themselves. But many others, including the makers of useful but niche-y iOS apps, clearly cannot. And while the royalty sought, which amounts to $5.75 on every $1,000 in sales, isn;t going to break anyone, companies have reason to object in principle to handing over payments to companies whose demands may or may not be valid, but which they cannot afford to challenge on their own.

Apple, which has so far been silent in the matter, should take the lead here and protect its community of developers. There are several things it could do. If it believed Lodysys’s claims are valid or aren;t worth the expense of a lengthy court fight, it should negotiate a broader license that would let it spread “pixie dust” over developers. At the same time, it should either persuade Lodsys to waive claims for past infringement or it should indemnify developers for any costs. Considering Apple’s financial condition, they could probably do this by collecting the change from sofas around One Infinite Loop.

How the BlackBerry Playbook Could Be Great

BlackBerry’s entry into the tablet races, the PlayBook, has certainly taken its knocks since it was introduced last week and most of them are deserved. No doubt, it’s an odd product. It’s a very business-y tablet in the world dominated by the consumer -centric iPad. And while I find comparing the raw number of apps available for different platforms a silly game, the Playbook is missing apps that any reasonable users would consider critical.

So I’m not going to recommend that anyone go out and buy a Playbook right now. And even once the most crucial missing apps are available, I would say the Playbook really only makes sense for people who already have BlackBerrys and use them for work. But for that not inconsiderable market, the Playbook could make a lot of sense.

Even the Playbook’s harshest critics have conceded some very nice design features: A fluid user interface that I find more intuitive than either iPad or Android. True multitasking (though it gets cranky when too many apps are open at once.) Speedy performance including a mobile Flash implementation that actually works well. Excellent 7″ display. Very good battery life.

The single most criticized thing about the Playbook is its lack of a native email and calendar functions. Instead, it is designed to pair with a BlackBerry and become, in effect, a synced display for the mail and calendar on the handset. and, in addition to a solid standard web browser, it offers a separate browser that works through the secured BlackBerry environment, a useful feature for deployment of corporate web apps. Once you accept the fact that only current BlackBerry users are the Playbook’s market, at least for now, this actually makes sense.

The FCC and Cell Boosters: A Cautionary Tale

The Federal Communications Commission has a full plate of complex issues including a nasty fight over the principles of network neutrality and a plan to repurpose unused or underused television spectrum for mobile data use. Anyone who expects a speedy resolution to these tough disputes should consider how long it has taken the FCC to resolve the vastly simpler issue of cell phone signal boosters.

On April 6, the FCC issued a Notice of Proposed Rulemaking (NPRM) on a plan to create regulations for the boosters, devices that amplify and retransmit wireless signals to allow calls and data transfers in areas of marginal reception. The notice will appear sometime soon in the Federal Register . Comments will be due 45 days after publication, responses to the comments 30 days after that, and then the commission should proceed to a vote and the regulations will be adopted. Don’t hold your breath.

One tip-off is that the NPRM runs 59 dense pages, including a three-page list of commenters on earlier, related proceedings. This fight dates back to at least November, 2007, with CTIA, the trade association of wireless communication companies, asked the FCC to ban cell boosters. The request was part of a petition to ban the sale of devices designed to jam wireless signals, and the conflation of the two issues has plagued the issue ever since.

Cell boosters are fairly simple devices. They can be mobile units that attach directly to a handset, car units that boost a signal within a vehicle, or fixed units to improve reception in a home or other building. They consist basically of a high-gain antenna, an amplifier, and a transmitter. Joe Banos, chief operating officer of Wilson Electronics, the leading maker of the products, admits that under some circumstances, booster, especially badly engineered ones, can cause problems for wireless network. The biggest issue is oscillation. A badly designed booster can pick up its own transmissions and rebroadcast them, creating the radio-frequency equivalent of the howl of audio feedback that occurs when a microphone gets too close to a loudspeaker. But the typical response to technical  problems of this sort has been technical standards and regulation, not a ban. “It’s like Carterfone,” says Banos, referring to the decision that opened up  the wired phone network to customer-supplied devices. “A booster looks to a cell site like a nearby phone.” The argument is that any device that plays by the rules and doesn;t harm the network should be permitted.

It’s not quite clear why the wireless carriers hate boosters so much.  Partly, it may be because they prefer to sell their own boosters, such as femtocells. These are min-cell sites that generate a wireless signal and connect to the phone network via the wired  internet. The problem is that they only work for fixed locations, not the mobile use that Wilson says accounts for 60% of sales. Partly it may be the desire of carriers, particularly Verizon wireless, which has been the most vehement opponent of boosters, to main control over everything on their networks.

The FCC didn’t grant CTIA’s request for a ban, but it didn’t create a procedure for approving the boosters either, leaving the devices in legal limbo. In November, 2009, Wilson requested that the FCC begin a formal rulemaking proceeding to set standards. In January, 2010 the commission asked for comments on the proposal, and 15 months later  formally began the proceeding with, of course, another request for comments. And it should be noted that this glacial progress is typical of how the FCC functions, no matter which party is in power of who the chair is.

“We’re also hopeful that the NPRM process can bring Wilson Electronics and the cellular service providers to the table to discuss what needs to be accomplished on a practical technical level,” Banos said in a statement after the FCC decision. “This would ensure that signal boosters can continue to provide individual and commercial users as well as government and public safety officials with a valuable tool that allows them to use their cellular devices in more places, with no risk of interference to providers’ cell sites.”

It’s just not likely to happen anytime soon.

 

 

Lessons from the “Samsung Keylogger”

On March 30, NetworkWorld.com published a report with the breathless headline: “Samsung installs keylogger on its laptop computers.” The next day, the story evaporated with an Emily Litella-style “never mind..” Along the way, though,  Samsung’s reputation was besmirched and NetworkWorld ended up looking foolish. Therte’s a lot we can learn from this sorry episode.

1. The original post (you have to find it among the updates; Google’s cached version has disappeared) should have raised a lot of red flags for NetworkWorld editors. Written by Mohamed Hassan of NetSec consulting Group with M.E. Kabay of Norwich University, it said flatly that Samsung had installed a commercial keylogger called StarLogger on a new R525 laptop. The authors failed to identify the software used to identify the keylogger. They also went on at great length gratuitously comparing the situation to a 2005 case where sony BMG distributed a music CD that installed a rootkit on PCs without ever explaining any relationship between the cases. The piece ended: “Samsung! We see a class action suit in your future.”
2. The piece failed to explain what possible motivation Samsung would have for installing a keylogger, Sony BMG, though its actions were stupid and illegal, at least had an anti-piracy motive.
3. There’s a real question of whether  Samsung was given a fair opportunity to respond. Hassan reported his findings to technical support and, after escalating the issue, was told the software had been installed “monitor the performance of the machine and to find out how it is being used.”  Assuming the accuracy of the report, the lesson is obvious. Corporate employees at any level should not offer uninformed opinions or speculation, especially on matters with potential legal implications. Kabay writes: “We contacted three public relations officers for Samsung for comment about this issue and gave them a week to send us their comments. No one from the company replied.” We need more detail on the nature of these efforts. If Samsung PR indeed blew the inquiries off, they share in the blame for what ensued.
4. The software responsible for finding the “keylogger” was eventually identified as Sunbelt GFI Security’s VIPRE. The identification was based on finding a directory called C:\Windows\SL on the machines. StarLogger creates such a directory Unfortunately, so does the obscure but entirely innocent Slovenian language support of Microsoft Windows Live.
5. To its immense credit, GFI came forward quickly to admit the mistake. In its GFI Labs blog, General Manager Alex Eckelberry explained how the error occurred and said: “We apologize to the author Mohamed Hassan, to Samsung, as well as any users who may have been affected by this false positive.” Eckelberry is right: False positives happen and they can cause a lot of grief. When one slips through, the best course is to fix it quickly and apologize.
6. Hassan and Kabay are both identified as security experts, but their expertise in this case seems to have been limited to running an antivirus program and unquestioningly accepting its findings. They present no evidence that they attempted to instrument the computer to find out whether a keylogger was collecting or sending out data. It’s not clear they ever looked inside the C:\Windows\SL directory.
7. NetworkWorld hasn’t apologized to anyone, which is inexcusable.