Forgot your password?
typodupeerror

+ - Highly respected engingeering school graduates more women than men 3

Submitted by kevmeister
kevmeister (979231) writes "Harvey Mudd College, a highly regarded engineering school in Claremont, California, announced that 56% of the latest graduating engineering class was female.

The article makes it clear that Harvey Mudd did put substantial effort into increasing female participation in STEM majors and that the overall graduating class or 2014 was almost half women.

Looks like (with effort) it is possible to get women interested in STEM."

Comment: Re:His choices... (Score 3, Interesting) 194

by kevmeister (#47351003) Attached to: The Internet's Own Boy

Information doesn't want anything. People want to be free.

While the famous quote is personifying information by implying will, I believe that the statement is effectively true. Nature has no free will, so it is not really true that "Something there is does not like a wall ", but entropy clearly demands that they fall and it looks to me like entropy wants information to be free, as well. It takes a great deal of effort to keep information captive, but almost no effort to release it.

People, on the other hand, purportedly want to be free. It takes serious effort to remain free. And, looking at support by the general public for "Big Brother" government (as long as it keeps us safe), it is not clear that most people even want to be free. :Or, perhaps they (or I) fail to understand what freedom really is.

Comment: Re:What is a gigawatt per hour? (Score 1) 461

by kevmeister (#47322213) Attached to: Half of Germany's Power Supplied By Solar, Briefly
Since electrical capacity or consumption is almost always measured in watt-hours, I strongly suspect that the number was 22 gigawatt-hours. Some editor "corrected" the '-' to a '/'. Of course it could be the author who simply misunderstood. To those who never made it to high school physics, gigawatts/hour sounds reasonable. After all, if you don't know what a watt is, you can't have a clue that there is a watt-hour. Most things are measured in something per something (e.g. km/hr or km/liter). Torque is the only hyphenated value most people ever see and few ever take the time to think about what any of them really mean.

Comment: Re:Redbox Instant (Score 2, Informative) 364

by kevmeister (#47196915) Attached to: Netflix Trash-Talks Verizon's Network; Verizon Threatens To Sue

I think you need to learn how routing protocols work. I will give you a hint, unless they are using 20+ year old protocols like RIP v1

I don't think you have worked for a real provider for quite a few years. RIP v1 (which is way more than 20 years old) has been effectively dead for years. Every provider has used either OSPF or ISIS for years. A few smaller providers may still use EIGRP, a pretty good proprietary protocol developed by Cisco. This goes back to at least the beginning of the commercialization of the Internet in the late 90s.

But these shortest-path protocols are only used for "interior" routing. That is, within a single administrative domain, like Verizon or Comcast or the University of California at Berkeley. (All of these entities actually have more than one administrative domain to make things manageable or to deal with organizational requirements.)

Between these domain a border protocol, BGPv4 is universal. It is also fairly stupid as it has no information on the interiors of the networks it is talking to. Instead it has a set of metrics that decide what routes to prefer and filter to control what routes are even accepted from neighbors (usually called 'peers'). There are very few metrics. the main one is called AS path length, or the number of administrative domains between points. It is fairly common to edit this path to make one or another path preferred, usually to prefer less expensive paths. Use the free or cheap path if you can and only use the relatively expensive path when there is no other choice. This is a gross over-simplification, but his i not a networking class.

The North American Network Operators Group (NANOG) has several excellent tutorials on routing protocols free on-line if you want to learn more, but, as simple s BGP is, the actual ways it is implemented get very arcane.

Netflix would probably be interested in evidence that Verizon is deliberately limiting traffic. I'm sure that they would be delighted to hear from anyone who can provide things like records of chats or e-mail where Verizon employees makes statements demonstrating this.

Comment: Re:didn't they decline H264 on Windows a while ago (Score 1) 403

WARNING: Revised history alert

Just a few short years ago Mozilla declined to support X.264 on ALL platforms event though there was a native plugin for Windows and open source support on other platforms. This was because H.264 uses a number of patented techniques and Mozilla wanted VP8, a patent-free codec.H.264 clearly won the war as every other browser supported H.264 for its HTM5 support. There was little support for VP8 (or, later, VP9).

Time passed and uptake on HTML5 using H.264 started growing in popularity. More and more pages failed to load properly on Firefox which increased the use of other browsers. Mozilla accepted the power of the market and added H.264 support to Firefox. Once the "standard" was written to allow DRM blobs, the handwriting was on the wall and Mozilla had learned the lesson well enough to at least provide a good, sandboxed way of supporting the blobs.

Do I like it? Hell, no! But I accept that most people simply don't care and it's either supporting DRM blobs or doing without and, while I might go with "do without", the vast majority will not.

Comment: Several errors in TFA (Score 1) 137

by kevmeister (#47034971) Attached to: Grace Hopper, UNIVAC, and the First Programming Language

While the section on Admiral Hopper looks correct to my knowledge, there were some hitorical flaws.

The UNIVAC I was produced after Remington Rand purchased EMCC, though Grace Hopper did work for EMCC prior to its acquisition a year after she started work there. The UNIVAC I was built by Remington Rand. Four years later, Remington Rand merged all three of their computer related operations into the UNIVAC division. The following year Remington Rand merged with Sperry to become Sperry Rand and the UNIVAC division was renamed as the Univac Division of Sperry Rand. Again, in 1986 Burroughs (another early office equipment company) merged with Sperry Rand to become Unisys. It is incorrect to state that Univac was "acquired" by Unisys as Unisys did not exist unto the merger of Sperry Rand and Burroughs. Wikipedia has what I believe to be a correct history of Univac.

The article also states that "Punch-card calculating machines already existed, but crucially, UNIVAC was programmable." I worked on IBM "Accounting Machines" and I assure you that they were programmable. See the article on the IBM 402 and 403. It was programmed by moving wires on a control card... similar to an old telephone switch board. The control board is pictured in the article. Programming was limited and painful, but it was certainly programmable and surprisingly powerful for its time.

While at Lawrence Livermore Laboratory I was fortunate to attend talk by Admiral Hopper (Ret.), then working for Digital. It was a great talk, but she didn't bring enough nanoseconds for the overflow audience, so I am sad to say that I don't have one, though I did have an RG-59 coax nanosecond I had made myself to explain why cable length was critical to certain synchronized operations.

Comment: Re:No link to opt-out in article? (Score 1) 85

by kevmeister (#46848203) Attached to: Verizon's Plan To Snoop On Its Customers

I went to the Verizon Wireless privacy link and both lines (my wife's and mine) were already opted out. It is very possible that I has previously heard of this and changed my settings, though it was not done recently.

I tried to go to the " autoads" page, but I found that to opt out, I had to enable both javascript (no surprise) and cookies. Also, the opt out is shown as a beta tool, so even if I allow cookies and javascript, who knows if it will actually do anything. Hmmm.

Comment: Re:What's wrong with that list = Heritage Foundati (Score 4, Insightful) 410

by kevmeister (#46264583) Attached to: Obama To Ask For $1 Billion Climate Change Fund

The whole idea of the grants is to get things going. While Tesla got most of its startup funding from Musk, most high-tech companies need money from either a venture capitalist or, if seen as too risky, some sort of grant.

In general, the funded firms companies were long-shots with a very significant up-sides for the nation and the economy, and the environment if they succeeded. It was completely expected that many would fail. If the odds were not long, most of the companies would have gotten private capitalization.

And, yes, several more may yet fail, but even the failures are far from a complete write-off. Some produced some potentially useful tech that could not be monetized before the cash ran out or the value was clear enough to get private funding.

Comment: Re:When I hear "I work 60 hours a week"... (Score 1) 717

by kevmeister (#46257613) Attached to: Your 60-Hour Work Week Is Not a Badge of Honor
Time passes and people forget the past (basically before their parents were adults).

Back at the turn of the century (1901, not 2001), according to the US Bureau of Labor Statistics the average work week was 53 hours and the "normal" work week was six 10 hour days. It really didn't change much until 1926 when Henry Ford reduced the work week at Ford Motor Co, to 40 hours, though the average work day had been slowly dropping from 10 to 8 hours before that time. But Ford was most likely the first major company to drop from a six to five day work week.

So, yes, people CAN work 60 hours a week and your grandparents (or great grandparents) probably did unless they were farmers. Farmers worked longer hours.

Comment: Re:MAC address, not IP address. (Score 1) 371

by kevmeister (#45921571) Attached to: Coca-Cola Reserves a Massive Range of MAC Addresses

Thank you for that. Very informative.

Do modern switches signal fake collisions to attached ethernet devices when their switch memory is exceeded?

No. 802.3 added flow control to the spec and that is the only standard method of dealing with this. It is optional and not all devices support it.

Faking collisions is not really practical as a collision is signaled from the transceiver, which used to be a separate physical device, to the MAC (Media Access Control) of the transmitting system. There is no mechanism on Ethernet to support sending a collision to a remote system. If running half-duplex (again, mostly obsolete), you could set your transmit line to block the transmitter, but that would trigger an error (collisions are not errors) that would disrupt the transmission and result in a packet being dropped. This would be way to disruptive, though dropping a packet due to lack of buffer space would be similarly disruptive.

This all gets into very complex issues as to much buffer space will break TCP congestion detection and severely impact performance. This is not for forum to discuss these issues, so I'll stop here, but you can search for buffer bloat" for lots of information.

I also highly recommend the use of the Netalyzer tool from ICIS.

Comment: Re:MAC address, not IP address. (Score 1) 371

by kevmeister (#45853721) Attached to: Coca-Cola Reserves a Massive Range of MAC Addresses

MAC addresses specify the backoff time for collisions on a LAN and aren't used at "worldwide" scales. They get stripped by the first router that sees them.

MAC addresses have nothing to do with collision back-off time. The back-off time is an algorithmically specified value that depends on the number of collisions (up to 16) that have occurred while attempting to transmit a frame and a random number. Collisions only occur on half-duplex Ethernet which is not normally used on modern LAN implementations. They are dropped (along with the entire frame header and CRC) by the first layer 3 (routing) device to process the frame, but are potentially used globally when an Ethernet frame is wrapped in another frame such as in some VLANs and increasingly popular Software Defined Networks (SDNs). Since they are guaranteed globally unique, they can be very handy for many things.

Only hardware vendors that need to provide unique collision avoidance characteristics on any customer's LAN need MAC address allocations.

Again, MAC addresses have noting to do with collisions and most LANs installed in the past decade have no collisions. They are used for addressing at layer 2 of the OSI reference model (Data-link). They are not actually a part of the Ethernet (IEEE 802.3) spec, but of a more global specification for creating globally unique hardware identifiers for network devices (IEEE 802) and are used by several LAN types which never had collisions (E.g. token ring, MAP (EtherBus), and FDDI).

As has already been pointed out, 16 million is the SMALLEST block of MAC addresses assignable, so this is far, far from massive. It does indicate that they plan on providing a globally unique ID for every machine which may or may not be actually used for addressing purposes.

Comment: Re:Clementine (Score 3, Informative) 317

by kevmeister (#45624547) Attached to: Ask Slashdot: Best FLOSS iTunes Replacement In 2013?
Rocket Player comes pretty close.It will allow my Android to do almost everything that an iPod will do including use the image from the file,though that has to be set in "Settings" or it will also use the image from the first song. The only place it fails is that it does not recognize the "Music Video" STIK.

Comment: Re:No question? (Score 2) 961

by kevmeister (#45585309) Attached to: Is the Porsche Carrera GT Too Dangerous?

For some reason nobody considers "I have no idea, we're still investigating" an appropriate answer.

Actually, almost everyone involved in some official capacity gives that answer. But that is not what the press wants, so they keep asking (Reporter: "Do you suppose speed could have been an issue?") until someone says something else that the reporter considers interesting (Tired cop: "It's possible") and uses that quote, or some mangled version of it. (Press: "Speed may have led to fatal accident").

I've been horribly misquoted by reporters and stopped talking to them about a decade ago.

Comment: Re: Why do people kiss-off KISS? (Score 1) 333

by kevmeister (#45275365) Attached to: How Kentucky Built the Country's Best ACA Exchange
OK. I'll agree thas CSS was a very good enhancement to basic HTML. It should, if used properly, actually save time and enhance reliability.

That said, I maintain a peronal site reporting local weather thst is pure HTML. The server for the single pahe is very basic... cat(1). It works and is very secure.

Comment: Why do people kiss-off KISS? (Score 1) 333

by kevmeister (#45270291) Attached to: How Kentucky Built the Country's Best ACA Exchange

Gee, isn't that the lesson of Google's initial success? Keep it simple and clean. No need for eye candy and extra bells and whistles. They add bugs and detract from the purpose of the site. As time has gone on, Google has drifted away from this concept, but the KISS principle remains valid and clearly had worked well for Kentucky.

Web sites, especially those with a single purpose, don't need 7 fonts, cool graphics, Flash flash, Java, or even CSS. Not that any of these (excepting Flash flash) are bad, but too many web designers seem to think that they are mandatory. The cost in development times, reliability, and ease of use is only justified by ego of developers and their managers. The purpose of the site is to provide important information and not to make health insurance look sexy!

The reason that every major university maintains a department of mathematics is that it's cheaper than institutionalizing all those people.

Working...