* Posts by BitterExScientist

17 publicly visible posts • joined 30 May 2016

Don't panic, friends, but the Chinese navy just nicked one of America's underwater drones

BitterExScientist

Re: Recon drone seized. Shocking?

@Big John

Best of all China will stop talking about how the US owes them money. Wouldn't want to risk someone falling into old habits and defaulting on some debt.

'I told him to cut it out' – Obama is convinced Putin's hackers swung the election for Trump

BitterExScientist

For some articles I just read the comments

And during the rest of the year they run Linux distro and Microsoft articles for the same reason.

For some articles, it's only the trolls I have respect for in the comments. Having lived in redder and bluer parts of TX and CA respectively, all I can say is that crazy doesn't discriminate. Atleast the nuts in the bay area are going to be too busy protesting Trump to mess with as many of the meaningful bits of local politics. Just let them have their resolutions and perhaps someone can finally fix a few potholes without weeks of community meetings.

Some! at! Yahoo! knew! about! mega-breach! as! early! as! 2014!

BitterExScientist

It! Hurts! To! Read! These! Titles!

Dear Reg.,

I know you're really excited about Yahoo, but please take it down a notch or use something less irritating like the blink tag.

Email security: We CAN fix the tech, but what about the humans?

BitterExScientist

Re: “Educate users not to open files that they are not expecting,” says Woodward

First quit training them to open up attachments and links in your emails the other 99% of the time.

IT and HR were the biggest offenders on my last job, and quite often to remind me to take security training, linking the online test and attaching the educational materials as a word file.

Deploying software every day is... actually... OK – what devs tell their real-life friends

BitterExScientist

High performing companies perform better than bad ones

Another "If we sort companies by performance, we find that the best ones perform far better than the worst ones using the the same metric ( eg power law distribution)" tautology based article?

How about making a nontrivial predictive statistical inference and showing other possibly confounding factors... For example, firmware, vs systems vs web are entirely different beasts.

No contract protected against the risk of bid-rigging, says expert

BitterExScientist

Tim sightings

He's still writing plenty of articles elsewhere. Given the publication in question, I'm surprised the editors didn't help him adjust his style. I really liked the detail and support he gave for his arguments, but they were structured like a mystery novel and I would have to go through half his article to figure out what point he was actually trying to make.

Space Data dudes say Google lifted Loon balloon tune

BitterExScientist

Please learn how to read patents.

As bad as all the X + internet ridiculously broad patents are that doesn't mean they all are.

This one doesn't remotely cover the broad concept of balloon networks nor the broader concept of balloon maneuvering as commentors here think. This one is narrowly written to cover a method to create such a network with broad stable network coverage by taking advantage of very high altitude weather behavior and is carefully drafted to stand up to challenge.

If Google's implementation fails to use any one part of the independent claim, they're perfectly free to do it. The dependant claims people are getting excited about place further restrictions on how the actual invention could be implemented.

You can safely launch your own balloon network without getting sued by them by for example staying below 40000 ft

BitterExScientist

Re: I thought you were being a bit sarcastic, but

@mongo

As much as I object to bs patents... I must be pedantic here.

You are looking at dependant claims and they should be read as giving specific versions of claim 2. The sillyness is to avoid someone else patenting claims 5 and 6 while accomplishing what your patent does and then suing you for the usage of your own patented invention no matter how obvious the additions were. Dependant claims often serve to illustrate how the invention is used. However, the independent claims are fair game.

Also, implementation details and how to create said invention are not put into the claims section...

Patent trolls, innovation and Brexit: What the FT won't tell you

BitterExScientist

Re: Few problems with this....

@patt attn

I'd argue it's just needing to get more than one rubber stamp, which is even worse for innovation.

1. Note that you say "based upon bad understanding of the technology" which means the result has little to do with a accurate assessment of the technology.

2. The patent attorneys I've always worked with said that if you get accepted immediately, you weren't nearly aggressive enough with the claims. A few rounds almost always got a patent passed, independent of examiner grumpyness.

So, in all, the process is a matter of getting several rubber stamps, and the expense only affect those who can't afford the billable hours (small businesses).

Bill Gates cooks up poultry recipe for Africans' paltry existence

BitterExScientist

First world problems: Lack of chickens

When traveling I always found that seeing chickens everywhere (and I do mean everywhere) was a good proxy that a country was not quite 'westernized'. The developing world has long ago figured out the value of chickens as an inexpensive source of protein.

Bill Gates must have some pretty good handlers for him not to notice during all of his trips. Real, non-factory chickens are fast and nearly impossible to catch or chase away. All I can imagine his plan doing is ruining the genetic stock of the local breeds into sickly flavorless creatures that can barely keep themselves alive creating a greater burden on the community. If people can't raise chickens in these areas it's a sign of something much worse going on that should be addressed instead.

BitterExScientist

Re: Chickens eat bugs

A real proper chicken (not a Western factory one) will also destroy a large vegetable garden given about 10 minutes and well before you can chase it out. Quite comical to watch as long as it's not your garden. Also after eating one you realize that most things don't actually taste like chicken.

Cisco axes unloved M-Series modular servers two years after launch

BitterExScientist

Re: Bingo!!

I think you can estimate the number of department heads involved in a project by the number of buzzwords in a PR statement; and in this case confirm the root cause of the failure.

Engine warning light appears on Uber's $100m driver settlement

BitterExScientist

Re: Labor protection laws in the the US?!

Worker protection no. Laws from multiple jurisdictions requiring lots of billable hours to say the equivalent of my self audit shows I'm compliant, no need to look behind the curtain; most definitely.

It's win win all around. Conservatives get to fundraise about the burden of government regulation, liberals about the "need to do something" (TM) to protect workers, and larger companies get to keep their labor costs low.

Anti-phishing most critical defence against rife CEO email fraud

BitterExScientist

The next big silicon valley discovery.

That has been well known for atleast the past 50+ years.

Administrative controls should only be considered effective against the hazards posed by lawsuits.

Or perhaps management already understands this quite well?

These big-name laptops are infested with security bugs – study

BitterExScientist

Re: Just load

... +1 As long as you're not making that suggestion to the manufacturers for their consumer PCs. Just imagine the enhancements they could do then, or look at the nonsense the carriers and manufacturers do to Android phones.

This is one case where I would wish Microsoft would be more of an industry bully, if it didn't seem like they're now aiming for these revenue sources as well.

Can I please pay money to receive a computer that already works and doesn't spy on me?

Rats revive phones-and-cancer scares

BitterExScientist

The Classics Work Best

This one is straight out of how to lie with statistics with academia's typical 'hot topic' modification.

Repeat the same study multiple times with a small sample set and eventually you'll hit pvalue = 0.05. Pick a hot topic and you don't have to repeat, just take your lottery ticket. You might not win but odds are someone will. News like this is why science has a reproducibility crisis.