Monthly Archives: September 2007

Amazon’s aStore is like Google’s CSE

Conceptually, I think Amazon’s aStore and Google’s Custom Search Engine are trying to achieve the same thing. Allowing the end users to customize and create manually edited, co-related pockets of well organized information.

Google’s CSE allows one to create a search engine that only searches among a subset of websites (web pages) making it more relevant for people searching in a specific area. In a way, it provides a way to create a vertical search engine for specific topic such as “vegitarian recipes”.

Similarly, Amazon’s aStore allows an affiliate to create a very specialized store. For example, here is an astore with books for learning indian languages.

Just like Google’s main website can provide end users to search for any information, Amazon’s website provides end users to search for any product. However, the aStores are manually edited and carefully picked so that they guide the users to concentrate on specific set of products.

Verticalization of any generic concept is always good.

Leave a comment

Filed under aStore, Google CSE

Float Behavior With Ads

When embedding contextual or non-contextual ads in a page, if you want to get the float behavior (where the ads are displayed with rest of the text on the page surrounding them both in the side and in the bottom), then it’s going to be tricky. Most of the ads are displayed in an iframe and further, the ads are displayed using a javascript and not by putting the iframe directly.

For example, this Notebooks page displays AuctionAds embedded within the surrounding context. The way to achieve this is to have

iframe {
display: inline;
float: right

in the main page. The css float concept is very useful to make the page estate well utilized.

Leave a comment

Filed under css

Gmail is loading and loading and loading and …

These days, I am really sick of Gmail. It’s getting too slow. Like a bloated pig. After all, the 1gb which eventually became 2gb strategy is probably not working. Having to mine all that piled up mails (which probably most people just kept it there, like me, out of laziness to remove or with the may be useful one day attitude) to provide some contextual ads (and a few times some coofeefool ad or the other which has absolutely no relevancy), it perhaps is taking a toll on the Gmail servers. Most of the night times and more on weekends I keep getting the “Loading” red button on the top. What I hate most about that is that it covers the signout link and as a result, I can’t even sign-out when it’s slow! Today is the first time since I signed up for Gmail, perhaps 2+ yrs, that I actually composed a notes on Yahoo’s email (yes, I usually keep notes in drafts and that’s me, don’t ask why I do it).

1 Comment

Filed under Gmail

News Papers, Blogs, Interviewers, Interview Candidates

Many graduating students are concerned about interview process and end up buying books, reading articles about what and what not to do while interviewing. But with blogging, the interviewing candidates equally get their opportunity to fairly evaluate the interviewers. Over the years, I probably interviewed more than a hundred people over phone and in person. Similarly, when I was graduating, I interviewed with a few companies. So, being on both sides of the equation, I would say I have a fair knowledge of what each side is looking for.

Anyway, I ended up at this specific blog about an experience interviewing with Google and I want to indicate a few paragraphs from this

“Anyways, doesn’t matter really, but this interviewer and I had a personality conflict and his accent was so thick I kept having to ask him to repeat himself. I think this is why they didn’t hire me.”

Yes, who cares how good a person is, if he or she is not able to articulate the problem clearly and coherently? On top of that, judging the candidate?

Next, take a look at this

“Well, I started to solve it and he kinda got annoyed. He got up and started trying to solve it for me (strange). Then he sat there for 3-5 minutes trying and eventually gave up without the solution. The best part was when he said, “just trust me”.

Yeah, right! Why not trust that interviewing candidate and give him a job :).

Anyway, the thing I want to highlight is, perhaps blogging gives the power to keep the equation more balanced in life. Just because one has the power to hire doesn’t mean he/she should act arrogantly.

In my personal experience of candidates interviewing for large corporations, below is a few key things that goes against them

1. Some times, they have to fly in from another timezone causing some physical exhaustion
2. Have to interview with multiple groups one after the other
3. All of this, sometimes in the middle of exams, middle of assignments etc and depending on how critical the school activities are at that time, half the mind might be the other side
4. Tell me which large corporation in the US doesn’t have people from different countries with different accents? For students who went to schools that have a lot of international students, they probably are already familiar with some of the different accents. Otherwise, it’s going to be tough.

There may be more reasons which you can think of!

Leave a comment

Filed under Uncategorized

For The First Time I wanted Spam Comments In My Blog!

Today WordPress had some maintenance to fix some issues. One of the issue that got fixed is the landing page when clicking on “Delete All” for the spam comments. Here is what used to happen for me in the past. It either used to give me 404 error and then later on used to go to the home page. Today, after the maintenance, when I tried it, it remained in the Comments tab itself. Finally! Well, now that I deleted all the comments, I wanted to verify that I wasn’t dreaming :). That’s when I wanted a bit of spam in my comments tab to be able to try it again! Funny, huh?

Leave a comment

Filed under Uncategorized

Oracle: Connect By Clause Performance (10g vs 11g)

I recently found that the connect by clause performance is much better in Oracle 11g compared to Oracle 10g. In 10g, the explain plan showed either a FULL TABLE SCAN or a INDEX FULL SCAN as the last step. However, in 11g, this is not the case. The number of consistent gets is much lower.

1 Comment

Filed under Oracle, performance

Performance Considerations For Mashups

Mashups are cool. Cool to develop. No denial. But what about the person accessing your mashup? Is their browser being hung up? Their bandwidth pounded with log of requests (be it AJAX or otherwise)?

A good mashup should pay attention to the usability aspect as well. Just pulling in information from lots of websites and assembling it on the fly on the client-side may not work. Especially if the information being fetched from the remote site has to be repeated for each row on a table of data, a standard implementation would definitely choke the browser.

Take Online Ad Networks page for example. It currently has mashups to get the Google’s PageRank and Alexa’s traffic rank. However, if this information is loaded on the fly for each row in the table, not only is it going to be painful for the viewer, it also would rapidly fire several requests to the respective servers from where the data is being fetched. So, instead, the information is loaded on demand, through the onmouseover events.

In general, in cases where a lot of requests have to be sent, instead of doing that by default, it would be good to do it on a user action. The choice is typically between a onclick and onmouseover. If you want to avoid the user to click too much, then go for onmouseover. However, if you want to absolutely make sure that the user actually intends to see the information you are providing, then an onclick would be better. Note that it’s still better than forcing the user to go to another page for the details and then bringing back to the summary page by back button or a bread crumb.

Leave a comment

Filed under mashup, performance

Is Amazon EC2 Right For SAAS?

I just ended up reading about how one company is leveraging Amazon EC2 for it’s hosted applications. That got me thinking on whether that’s the right thing. The proposition that there is no need for a serious admin is far too compelling when the salary for an admin who can manage the network, operating system, database and the middle tier is going to be quite high. But if you are already hiring an admin just to manage the rest of your infrastructure such as internal email, calendar, development environments etc, then may be that same person can do both?

Nevertheless, I want to talk about one interesting aspect of an enterprise SAAS application compared to a website. Say the SAAS is targeted to the US companies. In that case, the typical load on the system is going to be during the 8hr day. Keeping timezone into consideration, say it extends to 12hrs. The rest of the time, the load may be pretty much negligible.

But, just because the load is negligible, it’s won’t be possible to shutdown the instance. With Amazon’s EC2 stating

“Amazon EC2 reduces the time required to obtain and boot new server instances to minutes, allowing you to quickly scale capacity, both up and down, as your computing requirements change.”

would you tell your customers to wait for a few minutes before they can login into the system, should they choose to do some work from home later in the evening?

With the base computational cost of $0.10 an hr, it translates to $72 a month per instance. So, if you have two instances per customer, one for DB and the other for Middletier, that’s essentially a loss of $72 a month just to keep both instances running for that occasional load.

So, when someone is saving thousands of dollars to pay the admin, how does $72/month make a big difference? True. But it does translate to a direct cost to the company availing the SAAS. I don’t consider this necessarily a negative thing for Amazon EC2 or for SAAS vendors choosing EC2. I just thought of the obvious. Next, I need to do some more research on how Amazon’s S3 works with proprietary file formats like those of Oracle if someone is interested in running their database inside Amazon’s Web Service infrastructure. My understanding so far had been that S3 allows saving and retrieving files based on a key. If that’s the case, then it’s not really going to be possible to have proprietary files by databases. Does that limit having to keep the databases to the local disk of 160GB provided by EC2? I am not entirely sure.


Filed under Amazon EC2, amazon web services, SAAS

In the last 30 days, this house increased by $58K!

What? How? Why?

Well, I found on

* Zestimate: $843,569
* 30-day change: $58,598
* Value Range: $767,648 – $970,104
* Last updated: 09/11/2007

And then, looking at the neighborhood,

Past:        This
              home 	 94555  Fremont  Alameda  CA     US
30 days      7.5% 	-0.7% 	-0.1% 	-0.4% 	-1.5% 	0.4%
1 year 	      6.4% 	-0.4% 	-1.3% 	-1.5% 	-6.0% 	-5.6%

So, when the entire neighborhood is decreasing, why on earth did this house shoot up so much? No clue! I worked on enterprise business intelligence systems in the past and I know unless you can drill-down to the details, no one will trust the numbers.

1 Comment

Filed under Zillow