f00

There's also that little thing where Intuit and Citysearch fail to innovate and have terrible products. That just might have something to do with Yelp and Mint's success.

Funny, I assumed it was a thinly veiled jab at Mark Pincus.

I didn't attend SUS09, but by the video clips, it seems he committed this very "presentation sin."

I absolutely admit that OS X has a somewhat fuzzy appearance, particularly at small point sizes. This gets exaggerated with the MS Core/Web fonts, which were designed for the screen (stems that line up with pixel boundaries, etc), and have hinting that is specifically designed to be compatible with MS ClearType.

Ubuntu has similar (but not exactly the same) levels of fuzziness, but the rendering is inaccurate. Character tracking/spacing is the biggest issue, but there are others.

If you look at traditional fonts that are designed for print or general purpose use (e.g. Helvetica, Frutiger, Univers, Myriad, etc.), the accuracy and quality of OS X rendering stands out considerably. There is simply no comparison.. and once you get used to it, the small amount of "fuzziness" that you observe in small point sizes just doesn't bother you anymore.

1) OS X. A tad fuzzy, but everything is rendered _accurately_. Consistent thickness and character tracking (spacing).

2) Windows. Everything is slammed into pixel boundaries, making it look clean. Verdana is optimized for this style of rendering, so it doesn't look terrible.. but it's definitely not accurate.

3) Ubuntu. Worst of both worlds. Fuzzy and inaccurate.

Being fair to the parent, you don't seem to have the MS web fonts installed on Ubuntu, so the two OSes are rendering different fonts.

Still.. even if you do install the right fonts, the results will be the same.

Run, don't walk, to your nearest optometrist and get your eyes checked.

Ubuntu and OS X are not even remotely comparable from a typography standpoint. Using Freetype (the Linux/X11 lib for font rendering), tracking and stem thickness is inconsistent, ascenders and descenders get cut off, etc.

The Freetype team puts forth a valiant effort, but until they're allowed to use Adobe/Apple/etc.'s patented hinting algorithms, there's no comparison.

This line toward the bottom really caught my eye:

"Cselle said reMail will soon add a feature allowing users to upgrade by inviting 10 friends to to download the app."

This is a brilliant move.

I might have to borrow a page from Gabor's playbook very soon. :)

(Also, congrats on the new feature launch. ReMail looks great.)

If you're disappointed with the quality/performance of Fusion 2, why would you pay more and blindly upgrade to 3 without actually knowing that it has measurably improved?

Either way, please report back after the upgrade. The 'new features' list isn't terribly compelling, but a genuine performance boost (on, say, a dual-core MacBook Pro, not just an 8-core MacPro) would be.

Looking at your list, I see universities that have some amount of cachet or prestige, but not necessarily with specifically solid Engineering/CS programs. Further, many of them are in locations that are not traditional startup hubs (StL, Nashville, Philly, Cleveland, RI, etc.).

If your goal is truly focused completely on starting a tech startup after completing your degree, then you should be looking for:

1. Location. Go somewhere where you can find and meet like-minded people and the resources to help you get going. San Francisco, NYC, Boston, Chicago, maybe Austin. Cities with established startups and college towns are both great for this. PG conveys similar opinions in his essays about startup hubs.

2. A good CS department. You'll need fellow developers. Mine them from a good CS department.

3. Being in a school that is known to generate lots of startups is certainly helpful. Think Stanford, Berkeley, MIT.

Keep in mind, this is coming from someone who spent 3 semesters at Rice (before transferring to UT Austin).. It's a great school, but I would not choose it as a place to get a startup kicked off.

I'd also like to backup tsally's comments about large vs. small CS programs. This is the primary reason why I left Rice (although I was EE, not CS). The school is too small and the CS and EE programs did not offer the breadth of topics that the larger schools do.

> It's pretty clear that the core simulation objects have to be written in C++. C is too painful, and anything else is going to give an unacceptable hit in speed

The author lost me right here. C is too painful, and C++ is materially better? OOP is nice and all, but it's no panacea. I've worked plenty in both languages and find that both score roughly equally in 'pain quotient.'

Interestingly, the "LFSP" theory breaks down when you consider memory management and pointers. Those two features of "LFM's" cause far more difficulty for average developers, IMO, than the abstractions available in "LFSP's."

Disagree. Chris's point regards direct seed-stage investment from a VC, where a "pass" on funding an A-round indicates that the VC lost interest over time (suggesting that they might be disappointed in execution, think the market is no longer there, etc.).

My understanding is that Sequoia is simply providing more capital to YC, and that YC handles all the decision-making with regards to company selection and investment. For Sequoia to "pass" on investing in a YC-funded company after the seed stage is not the same situation.. they never "selected" the seed-stage funding recipient in the first place.

Edit: Regardless, this is all a silly thing to be concerned about. If your idea is even remotely viable, your execution is solid, and you have determination, you will find traction and success (and investment, if you really want it). If you're sweating things like this, you're getting distracted from the things that matter.

Yep, pretty much.

Just because preying on ignorance is easy and lucrative does not make it okay.

I can answer your question quite simply, and it's not about aesthetics or the magsafe power connector. The aesthetics are just icing on the cake. Very, very beautiful icing, but icing nonetheless.

I have better things to do with my life than dick around with xorg.conf for one more minute. I don't care whether the open source 'nv' or the closed source 'nvidia' module works better for my particular video setup, or what other voodoo incantations I need to do to get dual-monitor support working with or without compositing.

When I disconnect my laptop from my external display, I want to know that my window manager isn't going to strand my windows off on an unaccessible part of the desktop.

I want to know that when I put my laptop to sleep, that I'll be able to wake it up again. Quickly. Reliably. I don't need to think about ACPI or CPU power states, or why the kernel module for my wireless card won't work after waking up.

Linux works great for all my servers. I'm very happy with it, and I don't mind doing all the configuration work, because it makes a great platform for that purpose.

On the desktop, Linux is a non-starter. To quote Pulp Fiction -- "It ain't the same fuckin' ballpark, it ain't the same league, it ain't even the same fuckin' sport."

Bug with the most spectacular results:

As a (former) hardware engineer, I've worked on many projects where bugs have physical effects. This can range anywhere from amusing to seriously dangerous.

One such bug involved a mistake in the assembly diagram and silkscreen for a circuit board. The result was that a tantalum capacitor was installed backwards on a 12V supply rail.

Tantalum capacitors are polarized, and they fail in a spectacular way when reverse-biased. In this case, the supply rail could source upwards of 20A, so the fireworks were loud and impressive. Luckily the cap was easily replaced and the only permanent damage was cosmetic.

Hardest-to-troubleshoot bug:

In my subsequent return to the world of software, I worked on device drivers for network interfaces (among other things).

NICs frequently operate through a circularly-linked list of packet descriptors, which contain pointers to buffers in RAM where the NIC can DMA packet data. The hardware fills the DMA buffers and marks the descriptor as "used," and the driver chases the NIC around the ring, processing the packet data and marking the descriptors as free.

In testing, we discovered that under long periods (hours, usually) of heavy load, the system would occasionally freak out and stop processing packets. Sometime later, various software modules would crash.

Working backwards through the post-mortem data, I saw that the NIC would get "lost" and dump packet data all over system memory. I dumped the descriptor ring (tens-of-thousands of entries) and wrote some scripts to check it for consistency.

To make a very long story short, when the NIC was stormed with lots of 64b packets with no gaps, it would eventually screw up a DMA transfer and corrupt the "next" pointer in the descriptor ring. On the subsequent trip through the ring, the NIC would chase an errant pointer off into system memory and corrupt other system data structures.

Since hardware can DMA anywhere in RAM, the OS is powerless to stop it. The resulting errors can be ridiculously hard to track down and fix.

> C++ does not guarantee floating point operations happen the same every time...

Can you elaborate more on this? Are you implying that two identical calculations will return different results?

Requires further explanation. :)

Very, very cool.. and I love the name.

That said, I regularly test APIs using actual usernames and passwords via HTTP Basic Auth. It will be a cold day in hell before I toss my credentials in cleartext into a web form. :-)

I really like the syntax-highlighted, pretty-printed JSON output, though. Really handy.

..but a bitch ain't one. ;-)

(With apologies to Hova.. and HN. I couldn't help myself.)

You're probably right. I was feeling negative, because I had hoped that Capital Factory would provide seed funding for innovative development and help kindle some entrepreneurship among top-notch hacker types in Austin.

I'm not really sure what motivated them to pick a hodgepodge of me-too web properties (a few niche craigslist-ish things and a pet medical site).

Outside of Capital Factory, it seems like Austin Ventures and friends will continue to fund more of the same recycled enterprise network/virtualization/identity management software companies, being started by the same management teams over and over.

I really hope the next batch of CF startups is more interesting/innovative. Good luck with your venture, it looks promising, even if CF didn't care to fund it.

Austin, TX.

A YC-ish group called 'Capital Factory' recently began operating here, but if you look at the inaugural group of funded ventures, they're either:

  a) Not attracting Silicon Valley caliber talent/ideas/entrepreneurs

  b) Not looking for the same type of startups as YC.

Also, given that YC appears to be going Valley-only, perhaps the 'Another Boston option' should just be Boston, MA, right?

I haven't worked in a Windows-based environment in a while, but I do recall in the Fx 2.x days that it was significantly faster on Windows. Perhaps this is just a case of Mozilla catering to the largest audience.

If that's the case, then I applaud the effort, as getting the masses moved from IE[6/7/8] to Firefox 3.5 is certainly a more valuable effort than besting Safari performance on Mac OS X. There's just a big part of me that hopes they can achieve both. :-)

Argh. I want to love Firefox 3.5. Really, I do. Modern web standards. Faster. Smaller memory footprint. Native Aqua widgets (effing _finally_).

The problem is, compared to Chrome and Safari, it's a dog.

I've been running the nightly (Shiretoko) for weeks now, and, like everyone else, I'm hopelessly addicted to various extensions during heavy browsing sessions (development, etc).. but when I want to pop up a browser, check something quickly, and move on with my life, Safari is the browser of choice. The initial launch speed and JS performance are orders of magnitude faster.

Gecko is a great layout engine, but my gut feeling is that XUL and friends are holding it back. Agree? Disagree?