HNHacker News
TopNewBestAskShowJobs

kennyroo

450 karma · joined March 22, 2008

http://www.linkedin.com/in/kpate
submissionscomments
kennyroo··on Is releasing early ever a bad idea?
If it's a web site, I'd suggest that you launch the smallest high-quality product you can, and then tweak or add something every day until it's where you want it to be. You'll change track along the way, but that's OK because you'll do it based on real user behavior.

I launched Planaroo.com on June 30, and I wish I had launched it two months earlier. Many of my assumptions about how people would use the site were just plain wrong. I spent a long time working on features that few people use, and not enough time on features that a lot of people use.

If your site involves user-generated content, you have no idea how long it will take to ramp up. If your has lots of text, you just won't know how it will do in search for weeks after launch. Planaroo did very well in Google search almost immediately, but it still doesn't show up in Yahoo Search after 10 weeks.

kennyroo··on Is releasing early ever a bad idea?
Cuil screwed up by doing the full-court PR press before the product was ready. They could have launched the product in a more low-ley fashion without the backlash.
kennyroo··on Is Google a Media Company?
I think that Base, while interesting, was a harder product to understand (much less use) for average users.
kennyroo··on Is Google a Media Company?
Google's take rate on advertising is much higher on their own web pages, so it makes sense that they'd like to move the ratio toward Google-owned pages.

YouTube videos seem to pop up a lot on the first page of search results despite the near absence of content that might normally drive that kind of performance. Also, YouTube links often show up in the middle of the first page, which smells algorithmic. Perhaps the same kind of thing is happening with Knol entries?

So... questions for HC:

1) How much of the first page of search results can Google 'take' before people notice? (There's a reason they called it Knol instead of something like Googlepedia.)

2) If users notice, will they care as long as they find the info they're looking for?

kennyroo··on Is Wikipedia saturated? Yep For a year, at least.
I'm starting to re-think the whole Wikipedia thing.

I've been using the site for research related to my online travel site, and I've noticed a lot of facts and figures that are either outdated, slightly off-the-mark, or just plain wrong. Without an update schedule or paid staff, some of the content goes stale very quickly. I suspect this will worsen over time.

None would be glaringly obvious to a casual reader (e.g. the number of rooms at certain Las Vegas resorts is wrong, the current ownership status of some casinos is outdated), but God forbid that people take it for truth.

I updated a couple of entries myself, but I no longer trust it on the whole.

kennyroo··on What were you doing during the first dot boom?
Hiding out at relatively safe Disney.com after escaping Excite@Home just before the fall.
kennyroo··on News Corp. Q4: MySpace Profits Down
Anybody know what "new features and costs associated with the startup of new ventures" means?
kennyroo··on What's Really Killing Newspapers
http://www.windycitizen.com is really interesting. Someone will figure out how to make local work online, and this looks promising. Good luck!
kennyroo··on What's Really Killing Newspapers
Newspapers are dying because the half-life for headlines today is much, much shorter today than it was in the past and because other options are faster and free. Today's printed newspaper is by definition yesterday's news. Who wants to pay for old news? Imagine a web site that charged for yesterday's news. Wouldn't it fail miserably too?

The problem is that online news isn't really free -- it requires funding from the dying offline business because online news doesn't have a viable business model. In fact, a lot of people don't start with the news provider's site at all -- we start with news aggregators and then do quick dives into provider sites without leaving a trace of revenue in their pockets. This simply can't last. The offline business will go broke before the online business can be made workable at this rate.

Successful news providers in the future will need to start with a much cheaper cost structure to match a much lower revenue model. Current media companies may not be able to make that transition. None of them are moving fast enough, and most won't even acknowledge the simple reality that there is simply no future for the printed daily newspaper.

kennyroo··on Open Flash Chart Plugin for Ruby on Rails
Ah... it was news to me. Interweb reception isn't good in my cave, I guess. ;-)

Good point re: Google. They definitely have nice charts but I'm trying to not rely on Google for everything. I'm already using them for ads, maps, and AJAX libraries, but I don't feel comfortable putting all my eggs on one colorful Googly basket. That makes third-party solutions that I can host (like this one) especially attractive. Spread the love, and the risk...

kennyroo··on Open Flash Chart Plugin for Ruby on Rails
Very nice plugin. Thanks for posting this, and to the developers for making it available. This is a much-needed addition.
kennyroo··on A small startup responds to Amazon's S3 outage
You're absolutely right. That's the plan going forward.
kennyroo··on A small startup responds to Amazon's S3 outage
Not sure. Would love to know if such a tool exists.
kennyroo··on Test Drive of the Google Hosted Ajax Libraries
I started using this last week and have had a very positive experience so far.
kennyroo··on Ask YC: Do I have a chance?
See how far you can get without raising money, even if you have to work part-time on the side. Unless you're building something that requires a lot of storage, bandwidth, or people you can get a lot done without a lot of cash. Your location outside a tech hub won't be a factor until you need to start hiring experienced people or you need to raise hundreds of thousands of dollars. By that point you may know if the site is going to be successful or not, and you can move if you need to.
kennyroo··on Ask YC: can we collectively figure out how to neutralize eBay?
Pierre never patented the auction system. He didn't believe in them at the time.
kennyroo··on Ask YC: can we collectively figure out how to neutralize eBay?
Full disclosure: I'm a former eBay'er who spent 5 years at the company and had a good experience working there.

I don't agree that eBay is evil. It's just a big company behaving like all big companies eventually do. Still, I've given much thought to why no Western companies (Asia is a different issue) have been effective at competing with the company over the years.

I think there are three issues:

1) Trust and Safety is what eBay calls the never-ending cycle of identifying fraud, modeling it, and preventing it from happening in the future. It's a difficult, expensive, ever-evolving challenge that is far more complex than I realized before working at the company. Craigslist, which has recently seen a massive flood of spam on some of the more popular categories, is seeing how difficult it can be to keep a lid on this stuff. eBay is in a never-ending battle against criminals who are often very smart and highly motivated. New entrants in P2P payment systems have to deal with this problem to even get started. People think there is a lot of fraud on eBay, but they have no clue how much of it is killed before it sees the light of day. In many ways, eBay is like an immune system that can fight off infections it has seen before. It's the new stuff that gets through the system. Competitors won't have the benefit of eBay's decade-long immunity memory.

2) eBay spends millions on search marketing to maintain their position by keeping buyers coming to the site. eBay is, by most accounts, Google's largest ad buyer. It didn't used to be this way. There was a time when most users went directly to eBay.com to start their shopping, and that's still true of auction fanatics. But a large percentage of sales on eBay are now fixed price (or buy it now), and commodity shoppers increasingly start their shopping with a Google, Yahoo, or Live search. The cost of acquiring new fixed-price customers will be a major problem for anyone hoping to enter the general goods retail category.

3) It's still the best auction platform. The bottom line here is that no one has come up with anything fundamentally better than eBay's auctions. Personally, I think this is eBay's soft spot because the company is focused on fixed price goods and because they've owned this market for so long. The margin on auctions is very, very high, and someone could take them on. A better (more fun, cheaper) product would be a good place to start.

I agree with those who suggest that niche players will chip away at eBay's dominance, much like niche retailers at the mall chipped away at department stores over the years. If I was going to build an eBay competitor (and I'm not), I'd choose a single category of easily-shippable, high-resale-value items (e.g. SLR camera lenses or whatever) and create a better C2C trading system that kicks butt on that specific category.

kennyroo··on EBay Architecture
The author of the post referred to in the article is one of the brightest architects I've ever worked with. (Former eBay'er here.) Definitely worth reading to get his take on how eBay works, but the lessons are applicable in other contexts.
kennyroo··on Ask YC: What (Natural) Language Should I Learn?
I took Mandarin classes while working on a project in Shanghai in 2005, but made stunningly little progress and finally gave up. Then again, that might say more about me than it does about Mandarin. You definitely need a teacher, and consider focusing on the spoken language if most of your communication will be verbal.

Korean is useful because very few Westerners speak it. (There's a reason it's called the Hermit Kingdom). Very cool high-tech country with a too-low profile IMHO.

kennyroo··on Improving Type on the Web -- Removing Widows and Orphans
This type of thing, while certainly low on any developer's to-do list (including mine), does indeed make for a more pleasant reading experience with very little work.

Personally, I like what they're doing because they're doing it in regular XHTML and not resorting to alternative, non-open tools.

Some sites (MSNBC, for one) are using Flash-MX to create headlined to achieve a print-like look, which seems odd because MSNBC has no print counterpart.

It's refreshing to see the IHT and the NYT working so hard to do good things with native tools.

kennyroo··on Ask YC: Why doesn't Google use better technology to build its services (e.g., Flash)?
For my own selfish reasons, I'm pleased that Google doesn't use Flash for applications. Using XHTML/CSS/JS/etc. is great because other developers can inspect it, understand it, and go from there.

For practical purposes, I suspect Flash may be a pain in the ass to use on their scale. Debug, version control, localization, analytics, and more probably wouldn't play well with the rest of their platform and processes.

Also, Flash isn't supported on many wireless devices including the high-profile iPhone.

Future releases of HTML and JavsScript will handle some of what Flash does today. Perhaps it's just not necessary.

kennyroo··on DHH: Are you sure you want to be in San Francisco?
I wonder if DHH means SF proper, or if he's actually talking about the entire Bay Area.

The Peninsula (Palo Alto, San Mateo) and the South Bay (Mountain View, San Jose, Santa Clara, Sunnyvale) are home to the popular web sites used around the world.

The issues you would deal with living in SF differ pretty dramatically from the ones you'd deal with living anywhere else in the Bay Area. (Except the expense, which is bad everywhere.)

← PreviousPage 2 of 2