Busy Signal

Sorry about the lack of posts lately. I've been in the web service hell. I think the main problem with web service API is that ideal granuarity of service conflicts with ideal interface for the service.

RPC-style web service API is unsuitable for typical medium to large web applications because too many calls per page has to be made, resulting in unacceptable performance.

State transfer (aka REST) style web service API is typically chunkier (less calls) but message payload could get too big and the reduction of payload size could easily increase payload and client software complexity. What's the point of building a web service API if a complex platform-specific client has to be built to use the API?

While being tormenting, I've been taking my breaks in WoW, grinding through monsters. Oy. I think I was a mule in my previous life.


Dave rants and Scott ponders about walls. I agree that walls are not fair to everybody, particularly to those who find themselves not within the walls, and that increasing popularity of invite-only events and services is not healthy for the community as a whole.

But I think walls are like security blankets. It's not about reasons but fears and wants. It takes courage or recklessness to let them go. Frankly, I think it's foolish to let go of the blanket unless there is either a big carrot or stick hanging over one's head.

When in doubt, think like a pig.

Building Walls around Blogosphere

I am sure I'll get a lot of fleck for suggesting but I think we need to start building walls around blogosphere, to keep out spammers, pretenders, and freeloaders. Isn't this elitism? Yup. It is and I don't care that it is. I want good content inside the wall and all the trash thrown outside. Links going out may result in pulling blogs inside or pushing them outside. It's a dynamic social network for blogs, if you will.

In the end, this is where we are going except the wall will be the wall of ranks, ranks that push splogs to the bottom and the rest to the top. But I think it's better to know exactly where it is that we are incrementally heading toward.

Feuds Abound in East Asia

After Japanese PM visited a cemetery where Japanese war criminals are buried, some 200 Japanese politicians and assemblymen made the same trip. To China and Korea, such visits are equivalent to German PM visiting Hitler's grave. Considering that there are ongoing land and natural resource development disputes without any relief in sight, expect things to heat up considerably and possibly result in some small scale military 'accidents'.

Most of the actions will be between China and Japan though because Korea is currently in midst of an internal feud between freedom of speech and national security.

It began with a Korean college professor posting some inflamatory articles online which praised North Korea and trashed America. Conservatives roared, saying he should be arrested and charged for endangering national security. Liberals want his freedom of speech rights respected. Interesting backgrounds: both the current Korean President and the current secretary of justice used to be human rights lawyers. Current leader of the opposition party is the daughter of General Park Jung Hee, the original military dictator of modern Korea who got assasinated by his own man.

Fun stuff. My opinion is that the professor is not altogether there but it makes no sense to jail a man for being crazy unless he hurts someone. Let him say what he wants and let his thoughts infect others. Time to crush them is when they band together and actually do something violent. Given that past Korean governments frequently used fabricated national security incidents to remain in power, more tolerance is called for, not less.

Splog Number vs. Experience

Dave Sifry writes that splogs consist of only 4.6% of blogosphere. I am not in the business of tracking such numbers so I don't know whether the numbers are right or not. However, my splog experience is definitely not jiving with his numbers.

Ego search, as Dave calls it, is what I most often used to find out who linked to me. Is it ego searching when a company searches for blog posts containing the company, product, or executive names? Anyhow, it's definitely broken since anybody can inject name and keywords into their post, relevant or not.

URL search doesn't help much either since injecting a link to me in a post is just as easy as injecting a name or keyword.

I can't help wondering how Technorati determines whether a blog is fake or not because it takes more than a glance for me to tell whether a blog is fake or not. I can't tell who wrote the posts. Anybody can loot posts from elsewhere as a whole or partially and post it to their own blog. I think link statistics is not conclusive enough either.

What I am trying to say is this: don't throw numbers at me; give me the experience I used to have before all the spam blogs and aggregators started appearing.

Giving up on Blog Search Engines

I am giving up on blog search engines because their search result quality has dropped drastically in the past couple of months, thanks to blog spammers.

Blog spammers are using backlinking, name droping, ping spamming, comment spamming, post ripping, tag spamming, content obfuscating, and what ever means they can think of to give their contents, ads, some eyeball time. The funny thing is, they are still not being as sleazy as can be because spammer contents and links are still mostly honest. It won't be long before this age of spammer innocence passes.

Until much better solutions are hatched and delivered, I'll just hang up my blog searching hat on the rack. Til then, thanks for all the fish.


Technically, Ning is essentially a listing and linking plus skinning platform, simple yet flexible enough to meet quite a number of simple application needs. Functionality of Ning-based applications are not feature-based but people-based just like real world town markets.

I think Ning is a good thing because it enables users to create and experiment with services they need or ideas they had quickly and cheaply. As to their business plan, I have some doubts but I can also see several paths they can follow to reach the green acres. Execution wise, I think they need to improve on it greatly. Scalability? No big deal IMHO if the capital and the talents are there. Platform business? Not really. I think Ning is more of a crossroad town that could turn into a boom town.