freerobby

There is no way that the average purchase costs $2.50+ to deliver, even amortized over the lifetime of the book.

Amazon can charge whatever they want; I just think they should be transparent about what people's net pay will be. It's disingenuous to advertise that you take a 30% cut plus "delivery costs," and then mark up the "delivery costs" to over 25% of the average book price. Apple uses a similar paradigm (processing fees do not come from their 30%), but they only deduct the actual processing fees.

If you actually read the article you'll see that he addresses this and that it applies to wifi purchases.

this joke deserves more love.

If you can build a full-featured Heroku Management app for $10 worth of your time, then please email me because I want to hire you.

I've been using it for almost two years, so yes, it is maintained :)

I disagree with the idea that startups are inherently riskier. Sure you may only have 12-18 months of funding, but you also have much greater knowledge of the funding situation going in. Big companies have layoffs all the time, and as an employee you are much more likely to be kept in the dark about the circumstances leading up to them.

Minor clarification: it said he was a Canadian resident, not a citizen.

This post explains well the problems with Richard Cotton's argument: http://ammori.org/2011/12/31/sopapipa-copyright-bills-also-t...

Agree with this completely, nobody is more qualified to talk about piracy than the guy who went to open formats, anti-DRM, etc. and is still growing his publishing business year after year.

weird.

If anybody buys this, I'd appreciate knowing what this book covers that the basho Wiki/FastTrack doesn't.

I like the idea. Two implementation critiques:

1) Let recruiters see the value before making them pay (e.g. wait until people want to reach out to somebody).

2) Come up with payment terms that don't discourage use. You want people to pay you and then use your service religiously. Per-use pricing forces people to ration the extent that they use your service which gets in the way of maximizing the value you provide them.

The maintainer of the cookbook doesn't use haproxy, and would like a second opinion. If you use haproxy and think my commits are an improvement, can you please leave a comment saying so?

I was skeptical when I saw this, but I figured what the heck and gave it a try. Fixed two completely dead 18V DeWalt drill batteries - they now hold a charge for about 45 minutes of use (new they gave just under an hour).

Explanation (via http://ysuusy.com/easybatteryfix.html):

"Over time the crystals within a NiCd cell begin to combine and get larger. This causes them to loose capacitance and also break through the inner seals of the cell. Surging the NiCd cell using a higher voltage causes the crystals within the NiCd cell to shatter and become smaller. These smaller crystals have a greater surface area and therefore contain a greater capacitance for energy."

This answer will vary depending on what you are building. Web app? Language/framework?

As far as not dropping in-progress requests, most web sites have multiple servers with an haproxy instance that redirects among them. You can deploy to only one box at a time, and haproxy can direct visitors to the boxe(s) that aren't in the middle of being upgraded.

There is no generalized answer to your larger question though. Many frameworks and architectures have different deploying methodologies because of their unique nuances. If you share what you're building on, I'm sure many folks here can direct you towards some best practices/how-tos.

The wars, combined, only comprise about $1T of our deficit.

He was citing what the numbers actually were. The CBO's projections a year earlier were way off because they had overestimated GDP growth and their employment numbers proved to be overly optimistic.

I'm not persuaded by public polling on these issues for the very reason that you point out: the hypocrisy in their demands. The public super majorities that you speak of are also much too casual observers of the political process to understand what any of this means in isolation. Hell, over 60% of the public doesn't even _want_ the US to raise the debt ceiling (I forget the exact number cited in the press conference, but it was in the 60s). This is despite the fact that the leadership of both parties has acknowledged it needs to happen to avoid disaster.

I'm also not sure the cuts are as immediate as you say. I can't find any details on what is being negotiated (both leaderships and Obama have refused to give specifics), but on the revenues side for example, Obama said today that even the tax increases that are being talked about wouldn't come into effect until 2013. If they are talking about raising the retirement age or restructuring Medicare, I wouldn't rule out that it would take place over time or have a grandfathering period.

I agree that the debt ceiling and deficit reduction discussions should be separate, but the ship sailed on that a while ago, so all of the above is written from the perspective of dealing with this stuff in the present context.

I don't understand your last paragraph. Of course we should make good on what we can but if Medicare and/or Social Security (mostly the former) need reform to remain solvent, then why is it stupid to make the cuts that keep them solvent? A working less-than-ideal program is better than a not-working ideal program, no?

I agree that an email should have gone out (instead of just a blog post), but it's not at all clear to me what is wrong with Ferdowski's message. This blog post contains lots of shiny graphs but doesn't explain in any capacity what was wrong with the statement. Changing the color of words is not a cogent explanation.

This callout is absurd and misplaced. Entrepreneurs do not owe TechCrunch their scoop. So Caterina left Hunch 6 months before it was announced? Really? Well then guess what Mike - she had the info 6 months before you did. Why does Mike Arrington deserve right of first refusal on anything that goes to print? Call founders or don't call founders but don't cry foul just because a founder wants to control his or her own (in this case personal) story.

I don't like pay-per-use trials (or billing strategies) because it disincentivizes people from using your product. I want people to pay for my service, and then I want them to use it as much as possible.

How is this dead at 7 points? I agree he is too personal but his points are spot-on and he does a great job of correcting the overly-simplistic metaphors that were dominating the media coverage.

My MBP has a 120 GB solid state drive and a 1 TB media drive. I partitioned 120GB from the media drive to do Carbon Copy clones so that it has self-contained redundancy.

I do incremental backups on demand with Time Machine.

+1. Overall a good list, but I hugely disagree with recommending a laptop and no keyboard/mouse peripherals.

Well said.

Another benefit of open sourcing specific modules and libraries is that you can share your complementary code without exposing or commoditizing your core technology. Github is a perfect example of this - they open pieces that are helpful to other developers, but not enough that somebody could clone their site or compete with them.

Thanks (and karma) to you both for your perspectives.

We're temporarily down with no permanent damage. But we have no way to relay that info to our customers because everything, including our users' email addresses, is inaccessible. Big oversight that I will be fixing as soon as we're back up.

It feels absolutely terrible to only be able to assure customers that their data is safe _after_ they write a panicked email asking if their data is gone.

Agree completely. How much more proof could you possibly have that TechCrunch only cares about who's raised money?