Please create an account to participate in the Slashdot moderation system


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:There is one effect TFA omits ... (Score 1) 348

This is very interesting, I never knew this. I thought she had graduated from Princeton with a degree in EECS. If she was a lawyer, it is interesting to understand why Kleiner hired her in the first place. She was not hired as a lawyer, but as an entrepreneur.

The one other thing that boggles my mind is why the following obvious consideration has not been widely debated. To put it charitably, no-one would mistake Ellen Pao for a movie star or underwear model. She is a decidedly un-attractive woman whose brittle edge is obvious even in video snippets. How likely it is that men who have such a lot to lose would risk everything to have an affair with her? We are not talking about junior programmers, but guys with serious money and hardnosed business skills. The whole thing is completely bogus, and I am glad that the jury saw through her.

Comment: Re:Reminds of of something at a past job (Score 1) 764

OK, you should blame me for this, not the guy since I am quoting from the memory of years back. It is quite possible that he had done this the right way. The point of my comment were the names of the variables, not the implementation thereof. Jeez, I did not expect a code review here. Anyway, since the guy's implementation stood the test of time and earned the company several 10s of millions of dollars in sales, you have to assume that he knew what he was doing.

Comment: Reminds of of something at a past job (Score 1) 764

Worked for a company several years ago, where the lead programmer in a project (a superstar by all accounts and someone who was responsible for pretty much 25% of the company's codebase) was discovered to have used non-PC names for all his local variables. They were surprisingly apt, and hilarious, but definitely questionable in the current political environment.

For example, he would have code like this to represent the most significant and least significant words of a 32-byte unsigned int:

hungLo = word & 0xFFFF;

hungHi = (word & 0xFFFF0000) >> 16;

wellHungLo = word & 0xFF;

The problems started happening when the company grew and and some women were added to the team and took over the responsibility for the code; you can imagine the sort of complaints we got; ultimately we had to have a semi-major release whose only purpose was to sanitize the codebase.

Comment: Re:Intel chip better than Qualcomm? (Score 4, Informative) 77

by Magnus Pym (#49262331) Attached to: Intel Will Reportedly Land Apple As a Modem Chip Customer

This. The modem performance of the latest Qualcomm chips are pretty close to channel capacity and at least 3dB better than the best from either Intel's or MediaTeks according to reports I've seen. That means roughtly that the modem part of the phone spends double the energy to attain the same speeds, and under power-limited conditions would provide far less throughput.

It is pretty difficult to get perfect performance out of a cell modem, the underlying theory is pretty complex, and translating these complex algorithms into a practical working implementation is incredibly difficult. Neither Intel nor Mediatek know how to close the gap. Qualcomm is probably the only company in the world that has the knowhow and brainpower to do this.

But here is the thing: there was a time before the advent of the iPhone when consumers actually cared about the quality of the modem or underlying wireless technology. They don't any more. Or at least, not enough to vote with their wallets and do anything that the phone-makers care about. No-one who likes the iPhone will ever switch away to Samsung or similar because the cell modem is better... especially since they are on Wifi more than 50% of the time.

This is why Mediatek has become a unbelievable success in Asia and now biting into Qualcomm's profit at the high end. This is also why Intel (after struggling to make a viable cell modem for years and years) is finally getting a shot at the big time. And this spells really bad news for Qualcomm... another case of the market picking a demonstrably inferior product.

Comment: Re:FWIW don't download K-lite_Codec_Pack-10.9.5 (Score 1) 85

by Magnus Pym (#49150133) Attached to: VLC Gets First Major Cross-Platform Release

-> Chances are if VLC won't play it, the video is corrupt.

I'm sorry, but this is not the case; there are numerous formats that VLC does not support well, which other open source media players handle just fine. For example, mplayer supports real formats reasonably well. VLC has only partial support. VLC does not work well with files that have been spliced/joined with ffmpeg (stuttering, pausing, time-jumping). Mplayer works fine.

I'm a big fan of VLC, but saying that it is perfect helps nobody.

Comment: How do we know this is not parallel construction? (Score 4, Insightful) 129

by Magnus Pym (#49021669) Attached to: The Technologies That Betrayed Silk Road's Anonymity
This seems like a perfect use of parallel construction: figure out who he is by using illegal/secret technologies, and develop a plausible narrative of how legal methods were actually used. Maybe we are jumping too quickly to the "He was stupid" conclusion.

Comment: This reads like a hit piece (Score 3, Insightful) 222

by Magnus Pym (#48631529) Attached to: Marissa Mayer's Reinvention of Yahoo! Stumbles
I would not call myself a fan of Meyer, and her use of her relationship with Page to screw over her contemporaries (read this book) has really left a bad taste in my mouth. However this article reads like a hit piece. It looks like some activist investors are trying to get her to do things she does not want to do (the article suggests returning the money back to shareholders and firing all the engineers). They are attacking her personally and that stinks.

Comment: Re:Still no decent source browser integration (Score 1) 156

by Magnus Pym (#48207967) Attached to: GNU Emacs 24.4 Released Today

ECB has not been updated since 2009... and it was very slow & buggy when I used it last. It is based on cscope which has little support for C++/Java.

What I don't get is: commercial text editors like Visual Slickedit have had fabulous source browsing capabilities for more than 15 years. Another example is sublime text. Why is this not a priority for the emacs devs, whom I would assume are hardcore programmers?

Felson's Law: To steal ideas from one person is plagiarism; to steal from many is research.