Forgot your password?
typodupeerror

Comment: Re:iterative innovation (Score 1) 417

by Software Geek (#42761371) Attached to: Are There Any Real Inventors Left?

The wright brothers are a good example of iterative invention. They were not the first people to say "You know what would be cool? A flying machine!" That idea was not new. They were the first to make a working prototype. Most of the Wright brothers airplane was just an incremental development on bicycle technology, which was an incremental development on sewing machine technology, which was an incremental development on fire arms technology. They spent about a decade iteratively developing their wings and propellers, alternating between theory, wind tunnel, and working prototypes. With each iteration, they incrementally developed a more practical understanding of bernoulli's principal. The result of all this iteration was the wing and propeller shapes that we know today.

But no one was paying any attention to their decade of iterative work. So when they announced it, the popular conception of a flying machine went from science fiction to reality overnight.

Compare that to today. How many things in your life would have been considered science fiction twenty years ago? Just to name a few: The internet, tablet computers, hybrid cars, drones. I can't see how anyone could say we have stopped making big new inventions. Perhaps we have stopped ignoring the incremental steps that lead to any big new invention.

Comment: Software Patents Stifle Innovation (Score 4, Insightful) 249

by Software Geek (#39011339) Attached to: A Defense of Process Patents

it's a misconception that patents stifle innovation

The problem with software patents is that the legal standard of patent quality is too low. Something can be patented if it is novel and not obvious. I can write something that is novel and not obvious in an afternoon.

If I patent my afternoon's work, that doesn't enable me to sell it. So patents don't help me the innovate. On the other hand if tens of thousands of other people patent various things I could do in an afternoon, I am now legally obligated to respect all those patents, never doing any of those things unless I first negotiate a license with them. And I can get sued into bankruptcy if I make a mistake. And so the cost of avoiding infringing on low quality patents gets added to the cost of software development.

The patent system provides an incentive to be a software patent troll. It does not provide an incentive to develop software for people to use.

Comment: Business Model Confusion (Score 1) 321

by Software Geek (#38596076) Attached to: Why Freemium Doesn't Work

The author doesn't seem to understand the difference between the Freemium and Donationware business models.

In Donationware, you give away your product/service for free, and accept donations. Hopefully the donations bring in enough money...

In Freemium, you give away one product/service for free, and sell a second service. For example, you give away peanuts and sell beer. Hopefully the presence of free peanuts makes people drink more beer...

I tend to agree with the author that Donationware doesn't work very well. Most people have better places to give charity than to software developers.

Freemium can be more successful precisely because customers are not buying the upgraded service out of charity. They're getting real value in exchange for their payment. But in order to work, the free service and upgrade service should not be in competition with one another. If done right, the better the free service is, the more free customers are generated, and this eventually results in more paying customers.

The author's complaints seem to be that his customers complained about his peanuts, and didn't feel obligated to buy any beer. He should have made better peanuts, or just lived with the complaints. He should have made better beer, or found another business model.

Comment: Comedy of Errors (Score 2) 156

by Software Geek (#37297198) Attached to: Court Renders $3 Judgment Against Spamhaus

The defendant made legal errors regarding personal jurisdiction, thus losing their opportunity to get the case dismissed for lack of jurisdiction AND their opportunity to argue the merits.
The plaintiff made legal errors regarding discovery, thus losing their opportunity to recover damages.
The trial judge made legal errors that twice resulted in the verdict being overturned.

In the end, the case was decided without any actual evidence on the merits being admitted.

Justice!

Comment: Re:Read "The Party's Over" (Heinberg) (Score 2) 482

by Software Geek (#36968352) Attached to: Limits On Growth of Energy Use and Economies

The point about the assumption of growth is an important one. The world's financial systems are built on that assumption i.e. anyone who lends money expects to make a profit on the loan.

Actually, this was the one implication of TFA that I had a problem with. The profitability of an investment is not necessarily tied to growth. It is perfectly possible to use your resources to make tools, use the tools to make resources, and end up with more resources then you started with. Profit! While it is not possible to perpetually plow the profit back into the business and perpetually grow the profit, it is possible to perpetually plow the principal back into the business, and live off the profit.*

*For small values of "perpetual."

Comment: Re:Ridiculous study (Score 1) 482

by Software Geek (#36968242) Attached to: Limits On Growth of Energy Use and Economies

Suppose you have a bacteria colony that grows exponentially, doubling in size every day, until their resources are used up, and they all die.
Suppose this colony grows exponentially for a thousand years, doubling in size every day.
Fully half the bacteria are born on the last day before the resources are exhausted, and they all die.

Exponential growth is not pretty.

And yes, things do change quickly.

Comment: Chump Matching (Score 2) 355

by Software Geek (#36694626) Attached to: RIAA Math: Sell 1 Million Albums, Still Owe $500k

Your comment is based on the false assumption that the music business is about selling music. Since the artists provide the music, they are entitled to the lion's share of the profits.

In reality, the music business is more of a chump matching service. The music companies match up chumps who are willing to pay for music with chumps who are willing to sign away their rights to music. Since the music companies provide the chumps, they are entitled to the lion's share of the profits.

Comment: Getting in the Door (Score 0) 427

by Software Geek (#36644236) Attached to: Calling BS On Unpaid Internships

My experience in the software development field is that employers generally consider two kinds of software developers to be a "net negative:"
1) People less than two years out of college
2) People in their first few months at a new job.
They perceive these people as not understanding their trade and/or not understanding the job that needs to be done. They may be able to produce something of value, but they require a significant amount of mentoring by someone who is much more productive. Further, there is a significant chance that anything actually produced by an inexperienced developer will have to be thrown out and/or redone.

This perception makes it hard to enter the field. Employers much prefer to hire someone with a few years of experience rather than a new grad.

All the "If the company asks you to work for free, they are trying to rip you off" comments miss a fundamental point: If interns were economically advantageous for the company, there would be a lot of companies staffed mostly by interns. But there aren't. Instead you see companies very reluctantly get one or two interns in order to address their corporate citizenship obligations. And then they neglect the interns, because they don't care enough about corporate citizenship to actually invest in the interns.

Co-ops and internships (paid or unpaid) are a way for newcomers to the field to get past this barrier to entry. I'm not saying every intern has a good experience. But internships in general address a very real problem for the intern. My advice to anyone who wants to be a software developer is: get as much on-the-job experience (paid or unpaid) as you can while you are in school. If you don't, you will have a difficult time finding a job at graduation. There is a very real chance that you will have to take a job that diverts your career in a direction you don't want it to go.

Comment: Users vs Programs (Score 2) 153

by Software Geek (#36509658) Attached to: PlanetLab Creates a More Advanced Sudo

The problem with the Unix security model is that it is designed to protect users against other malicious users. It does this by allowing each user to trash his own space, but not anyone else's space. But in modern computing environments, there is usually only one user, and sometimes less, and the challenge is to protect the computer against malicious programs. So, letting every program trash the one user's space isn't really that useful.
Of course the Unix security model can be adapted to protect against malicious programs. But in practice it is so difficult that no one bothers to try.

It appears to me, after a brief scan of TFA, that vsys just provides finer granularity without addressing the fact that the security model is fundamentally broken.

We need a model that makes it natural and easy to run every program in its own sandbox.

Comment: Unintended Consequences (Score 1) 363

by Software Geek (#36058874) Attached to: Google/Facebook: Do-Not-Track Threatens CA Economy

"Do not track" is a nice way of saying "It is illegal to have certain kinds of information on a computer."
This makes me very nervous.
If my browser caches pages, does that mean I am illegally tracking?
If my web site saves logs for troubleshooting, am I illegally tracking?
If I use peer-to-peer file sharing software, am I illegally tracking?

What powers of search and seizure will the state need to enforce "Do not track?"

I haven't read this particular bill. Maybe it has some well thought out way of targeting advertisers without placing a huge new regulatory burden on everyone who uses a computer in California, or makes software that might be used on a computer in California. But I doubt it.

Comment: Re:Bankrupt government funds boondoggles (Score 1) 529

by Software Geek (#35926314) Attached to: US Funding Five Game-Changing Energy Projects

Since we have no shortage of energy but we have a desperate shortage of funds in the Treasury, these types of projects should not be funded. Let a less bankrupt country fund them.

These types of grants tend to be direct monetary payback for political support and campaign donations anyway.

To address your points one by one:
1) Bankruptcy
Cutting this research would reduce the federal budget by about 0.0025%
The United States has a serious decision to make whether to: Cut military spending, Cut Medicare spending, Raise taxes, and/or Go bankrupt.
No other budgetary changes that the US can make will make any difference to the crisis we find ourselves in.
Trying to solve our budget problems by cutting research spending is just a way of denying the real problem.
http://en.wikipedia.org/wiki/Amdahl's_law

2) No shortage of energy
The resources used by our current energy technologies will be exhausted in a few decades. All of the replacement technologies being considered will take at least a few decades to deploy on the required scale.
Energy is THE scarce commodity in our economic system. A discovery that halved the price of energy would double the standard of living of every person on earth.

3) Corruption/Inefficiency of government funded research
If government does not fund this research, who will? Private companies tend to fund R&D that will make them rich, not basic research that will make future generations rich.
Is it better to do the research inefficiently or not do it at all?

The following discoveries were brought to you by US government funded research:
Nuclear power, Communication Satellites, The Internet

Comment: Margins, not units sold (Score 1) 210

by Software Geek (#35794328) Attached to: Amazon To Offer Ad-Supported Kindle

How many more kindles will this pricing option sell?
Are there really that many people out there that would walk away from a $139 kindle, but are willing to buy a $114 kindle with ads?
The discount would have to be much steeper in order to result in a measurable increase in sales.

I suspect that this is really about improving margins, not selling units. People who were already going to buy a kindle will opt for the cheaper one. The ad revenue it brings in will more than offset the purchase discount.

Comment: Re:What year is it? (Score 3, Funny) 288

by Software Geek (#35633332) Attached to: MySql.com Hacked With Sql Injection

When interviewing people for QA positions, I routinely ask "Do you know what an SQL injection attack is?"
I have never yet interviewed a candidate who answered yes.
So, then I explain what an SQL injection attack is, and ask how they would test for vulnerability to one.
Almost without exception, the answer is "I guess I would try entering some special characters and keywords into the GUI, and see what happens."

We warn the reader in advance that the proof presented here depends on a clever but highly unmotivated trick. -- Howard Anton, "Elementary Linear Algebra"

Working...