jaylevitt

No.. having founded a company is a decision made and re-made by historians.

Steve Case was VP of Marketing when AOL was founded as Quantum Computer Services. As his role grew, early employees left, and time passed, he became the co-founder.

If you're writing code in any environment but an early-stage startup, the qualities it has when read by a human are ALSO the primary criteria.

Yep. I'm reminded of Don't Ask, Don't Tell, which in 1993 was a GIANT step forward for gay rights (you should have seen the elation during the March on Washington). It worked so well, it moved the Overton window so far, that 19 years later it represented all the facets of the discriminatory policies it had once replaced.

Sure, and Rails offers a similar syntax:

  User.select('x').where('name like ?', username)

But if your language allows literal string interpolation (as Ruby does), what prevents you from doing this:

query conn ("select x from users where name like #{username}")

How do type-safe languages prevent this?

I'm admittedly ignorant of any type system newer than C++. In a modern static language, how would you design Query such that any SQL injection is caught at compile-time?

On the dynamic side, Rails (in Ruby) doesn't currently catch SQL injections, but it does catch HTML-escaping injections. It (roughly) tags all strings as tainted by default, and when you send them to the browser, it escapes them. If you want to send literal ampersands, angle brackets, etc., you have to mark them as explicitly safe. Since most of your literal HTML is generated by templates (which themselves distinguish variables from static HTML), you end up with run-time safety unless you actively try to break out of it.

> He said "coke tasted better than pepsi"

I think he actually said "coke tastes better than pepsi". That verb tense has very different implications.

But isn't it problematic that it compared real-world average unit tests with the best-available type system?

Of course you have access to historical data; you're on the Internet!

And you're right - it's estimated to have dropped 13-15% in the past 30 years.

http://www.awwa.org/files/Resources/Waterwiser/JAW0211rockaw...

Their Returns Department charges a 15% restocking fee, return shipping unless you ask for a prepaid label, and original shipping unless you raise a fuss. What do you consider bottom notch?

Now you've heard of anybody else. I didn't have a horrible, traumatic, life-scarring experience, but I did have a bad one:

* They advertised an item as a SAS/SATA RAID cage; I paid rush overnight shipping, only to discover it was actually SATA-only

* When I tried to return it, I was charged a $90 "restocking" fee for their mistake. Indeed, their invoices state "All refunds require a 15% restocking fee." That's not legal in Massachusetts.

* I was also left responsible for paying my own return shipping (until I complained). Not legal either.

* And the original shipping, for an item that's not as advertised. Probably not legal either.

* When they corrected all this, they gave me the "one time exception" bit - which implies that their policy is to charge me shipping even if they send me the wrong item.

Color me not very impressed - and in fact I've avoided buying from Newegg for over four years now.

https://gist.github.com/2927900

Wouldn't emergency brakes pretty much kill you if you're traveling a few thousand mph at the time?

In some US states, too; many of our states have a code modeled on the Uniform Commercial Code. In Massachusetts, at least, all items come with an implied warranty of merchantability and fitness for purpose, e.g. that it's worth selling and useful. Sellers cannot legally waive those warranties.

As a buyer, my only responsibility is to allow the seller to pick up the item from me. If some site tries to pull a New Egg on me, I point them to the relevant laws, and make a very gracious offer - not only will I allow them to pick the item up, I'll even pack it all up for them in a box.

Lesson: Check your local laws before assuming that these "no returns" or "restocking fee" policies have any weight.

Conversely... I know a guy who worked with a well-known startup founder, but he didn't list that founder as a reference, because he was sure he'd burned his bridges. Just so happened that the hiring manager had the founder on speed-dial. Later that night, the manager calls the founder, who says: "If I were to start a new company right now, that's the guy I'd want with me."

You never know.

If Swype was obvious in 2003, why weren't there multiple (or any?) slide-to-type QWERTY keyboards written for PalmOS or early Windows Mobile?

I've owned PDAs since the Palm 5000 in early 1996, and I tried all the alternative keyboards, from Fitaly and Atomik to MessagEase and.. something else like MessagEase. I don't remember anyone even suggesting the idea of a sliding QWERTY keyboard; there were alternate-layout tapping keyboards, and there were star-layout sliding keyboards, and those were the two sides of the debate. No?

It's difficult to comparison-shop for features; Amazon reviews are now the definitive source of consumer opinion, but you have to sort of agglomerate your own set of W-X, X-Y, Y-Z and X-Z comparisons to see why people chose one over the other (and account for the fact that all four of those reviews were on models that have since been replaced).

Doesn't any 22-year-old have 22 years of computer experience?

Random fun fact: Sanford Wallace once crashed AOL's mail server by sending mail that was allegedly from "howmuchwoodcouldawoodchuckchuckifawoodchuckcouldchuckwood.com".

And then there was the Commodore 1670, a 300/1200 bps modem with auto-answer. Not only "with" auto-answer; with auto-answer on by DEFAULT, with settings in RAM, and with a computer that you restarted by hitting the power switch. In an era when most people had only one phone line.

A critical Commodore skill was the ability to respond to the 1670's answer tone by whistling a 300 baud originator tone, so the 1670 would "detect" carrier, then "detect" carrier drop and hang up the line. So you could talk.

And if you don't report back, we'll still have a data point..

The confusion probably stems from the fact that the product manufacturer, Musion, repeatedly claims that they build "3D Holographic Projections", even though they're obviously neither 3D nor holographic:

http://www.musion.co.uk/

This is the same company that did the Tupac "hologram". It's all based on Pepper's Ghost, a 19th-century illusion involving a projector with a screen that reflects the projection, but looks transparent in front of the actors.

Even if you were able to talk directly to the manager, you'd have been stuck. (I talked to a woman who claimed to be "manager of customer care and sales in North America", and she didn't even know how to work their systems.)

There seems to be no support management team in any normal sense, and no upper-management concern about the lack of support management.

Here's 59 ways they failed to deliver me a working product:

http://consumerist.com/2008/03/adobe-needs-eight-employees-t...

As the self-appointed permanent apologist for all things AOL, I declare that what Eric did is totally cool. Any work resources he used can be offset by our collective 24-hour work days from 1995-2000 - and we were denied Starbucks, wi-fi and telecommuting on account of they didn't exist yet.

I spent plenty of nights at the terminal and afternoons asleep on those couches; Eric's nighttime efforts will ensure that they wear evenly.

Eric: Welcome!

(AOL 1989-2001)

No. It wasn't a committee. It was specifically the new, MBA-credentialed "product marketing" person.

No, but (to me at least) it's MBA thinking. Technologists were thinking "How can we build something cool?" (An interactive mouse-driven sketchpad - over dialup in 1989.) Product folks were thinking "How can we build something great?" (What if we connected the whole world via e-mail and IM?)

MBA's were thinking "How can we maximize short-term revenue?" Modal popup ads!

I know how to mix live audio. Why can't I have a mixing board on my conference call?

"When I began my career in the late 80s and early 90s, MBAs ran most of the best new technology companies."

Not exactly. By 1993, MBAs were already starting to turn AOL into the "how much can we milk you before you'll walk away?" behemoth it became. They'd tried to weasel out of our promise of a "lifetime 20% discount" for charter members, they'd rounded up in some directions and down in others, and they'd literally removed the "Sign Off" button from the toolbar because we billed by the hour, and --

Well, I'll let the MBA tell you.

"We needed to drop one icon because of space/look and feel and it made sense to us out of all the other functionality to drop exit. In addition, some folks were uncomfortable with how visible the ability to get out of the service was--all in keeping with our philosophy to keep them online for that one additional minute :)"

That would be wonderful. Unfortunately, IdentityMap was disabled by default in 3.1 because it was incomplete, and it was pulled from 4.0 for a while due to lack of development:

https://github.com/rails/rails/pull/5261

Though there is now a company volunteering to fund completion of the feature, starting Real Soon Now:

https://github.com/rails/rails/issues/5442

IdentityMap would, obviously, be great. Though I didn't realize it could serve as a replacement for model caching; didn't IdentityMap, like the query cache, only last for the lifetime of a request? Or could it be persisted in the Rails.cache of your choice for an arbitrary lifespan?

There used to be many model-caching plugins for Rails, but few if any made the transition to Rails 3. The record-cache gem looks interesting and flexible, though I haven't tried it:

https://github.com/orslumen/record-cache

Maybe for his litigation skills? I'm sure there's a large legal team behind the scenes that includes IP experts, but "litigation" is its own field - most lawyers never see the inside of a courtroom. There's rhetoric, witness examination, making motions and objections, and so much more than expertise in an area of law.

(IANAL but I helped AOL sue Sanford Wallace, and it was a sight to behold.)

For what it's worth, AOL had this 25 years ago and never even had complaints. And yes, people came to realize that "DanBlake read this" meant DanBlake's client had read this, and you, personally, might not have seen it yet - or, if you had, you'd marked it unread for later reading.

But we had that feature from day 1; user expectations grew up around it. Introducing this kind of behavior mid-game is dangerous for all the reasons everyone's mentioned.