SezWho.com Homepage

Comments?

Some of the new things are:

- Site images

- Twitter integration (still TBD)

- More neutral scheme

- Efficient layout

What do you think? If you think you can do better, annotate this image and post it back here…Who knows, if we like it, we might even do a surprise

After our outage last week. Here’s a comeback on the events.

We found that our servers did not go down; the apache servers did (albeit incompletely without releasing the socket/port) This made the browser to wait forever to connect to the webserver, eventually timing out. This delayed the loading of sites which are part of SezWho community.

To add to this hubbub; the night before, our SMS quota on the monitoring service ran out and as such we did not get an immediate notification (One of our developers out of Bangalore, caught this at around 4am his time). We could have resolved this problem in less than 10 mins if the alerts came as usual. The SMS quota has been set to infinity now

This week, we got our new environment setup with bigger irons and we have started moving a few things there. We’re setting it up in a way to have the browser request time out immediately if the servers are not available. We are also exploring suggestions of externalizing the CSS.

We take the quality of our service very seriously and will be up and will do everything in our power to make sure an outage like this never happens again. We are thankful to our community members for their support, cheers (and jeers!)

We had a kernel httpd crash (Linux FC7) and our site went out for 3 hours around noon time. What can I say, ‘am feeling really sad about this. We just signed up with a lot of new users — I’m sure we made them very very unhappy

This was worst ever timing for us – We signed up with our new data center sometime ago and brand new servers got provisioned today (The actual move into the new facility along with the new architecture was planned next Saturday).

I cringe and squirm at the sight of my mail box, overflowing with bits of information, some useful, some actionable, some from customers and some from people seeking money. Same story on the web, overload of search results, over flowing memes — How do I find that content nugget which is of importance to me? and which is important to me right now?

We bookmark (as in delicious) and seldom go back; enable saving of web surf history but Google does not augment the results; a lot of productivity boosting applications are still experimental. The big question is how to fix the overload? Good that we have information at least, but how do we find that one bit of information which is useful to me at this second.

This week Google, Microsoft, Symantec, IBM, etc. met for their first ever Information Overload Research Group (IORG Forum). Their mission is grassroots awareness and a vision/architecture of providing solutions for reducing the overload. I’m sure the members companies are interested in rolling out the tools but having them sit and think about this is a good step in itself. Their mission reads:

We work together to build awareness of the world’s greatest challenge to productivity, conduct research, help define best practices, contribute to the creation of solutions, share information and resources, offer guidance and facilitation, and help make the business case for fighting information overload.

This is very interesting. Like everybody else we have a deep interest in reducing the information overload of netizens and increasing everybody’s productivity. With 281 Exabytes of data being created, just in 2007, I’m sure this is big enough to get everybody’s attention.

Yesterday was hectic — We hung out with Entrecard community (ate same some virtual Pita & chips while we were there), and saw some conversations coming from SocialSpark & CreativeWeblogging bloggers.

We saw good amount of traffic coming in — new users, new issues, new ideas and some fresh perspectives. For the code rollout which happened yesterday, we added some ‘happy’ features and inadverently added some unhappy bugs. We ignored Joel Spolsky’s Step #5. (Though, I wanna put the blame on the loss of quietness; you know, tweets, phone calls Skype, FriendFeed, Google Reader, Starbucks, and interruptions for lunch :))

To the list, here are the issues we found (most of them fixed):

• Layout issues when using custom UI (or custom CSS). While in the Entrecard chat room I advised folks to stay away from custom UI, rescinding that, this is FIXED now
• Blogspot script snafu. Due to a bug, we were just sending a key and not the scriptlet. This was happening to only a segment. This is FIXED now. We’ve resend the emails with the scriptlets to the affected users
• Profile Image upload. This was a technical issue where our local Linux system clocks were out of sync with Amazon S3 server clock. S3 allows a delta of 15 minutes. We found that our NTP server got misconfigured and was reading the wrong time zones. This is FIXED now
• WordPress 2.6. We were taking it easy on this release as it was supposed to come out early next month. To our surprise the release came out a month ahead of schedule. We are releasing a beta version of SezWho plugin for WP 2.6 (Download here)
• Blogspot latency. Our blogspot plugin has been slow in syncing up the comments and posts. We’re still working on this. Please bear with us if your published comment/post does not show up with the SezWho functionality right away. IN PROGRESS

Here are the features we’re going to work on:

• Support for multiple blogs in the same database
• Filtering out comments from the same site. Right now, you have to manually suppress them one at a time
• Merging of accounts
• Easier configuration of custom UI components like rating throbbers, stars, and boxes
• Rating layout issues. There are some themes where the layout gets all skewed up

If not for the issues, we did wanna sing some Beatles, today, but we are reserving that for Friday evening. A big thank you to everybody for their comments, conversation, and participation!

We are making 3 big announcements today

1. EntreCard: Entrecard is a blogger-to-blogger widget advertising network, has formed a partnership with SezWho that allows members to earn advertising dollars for their best comments. Below is the take of one of our common users, Lee, that summarizes things nicely:

Also see Graham’s post here.

2. Creative Weblogging: Creative Weblogging is one of the fastest growing international professional blog media networks, that is partnering with SezWho to bring reputation-driven user profiles to members. SezWho drives community engagement and traffic for bloggers by displaying their universal profiles with their contributions across SezWho-enabled social media sites. Users get the benefits of universal profiles, ratings, and comments, without giving up control of their data.

3. SocialSpark: SocialSpark is the world’s largest social media marketing network that will offer SezWho to help bloggers build and monetize their reputations based on their contributions and areas of expertise.

Welcome to the community everybody…and don’t be a stranger.

@, the at sign, started it’s career as a favorite amongst accountants. Then, in 1982 David Crocker wrote RFC 822, which defined that the lexical symbol “@” be present to separate the mailbox from the domain viz. person@domain.

During the 1.0 days, using an @ in a corporate brochure conveyed it’s online presence. To a large part the usage of @ was still in emails, until twitter was born 2 years ago. twitter’s @reply feature took away the symbol’s pull from the email. Now, we are seeing the people are increasingly using @ outside of twitter for calling out a conversation when threading is absent in a commenting system.

Take for example this slice of conversation from a blog:

The commenter in the above example _wants_ to talk directly to the other commenter except that the comments are laid out in single-dimension reverse chronological order.

We are noticing this new social behavior (mostly used by twitter users) where @ is being used to point to the conversation participants (eg. @anon_guy, I agree with @20, etc.). This is very popular practice amongst commenters in heavily commented blogs esp. on controversial topics where discussions sway in multiple directions. Is this a new social behavior? Does this new social behavior calls for a change in the way we see comments in the blogs? Is threading a requirement? Does threading invite more participation and 1:1 discussion between commenters within the realm of the larger topic which is the post itself? These are some interesting questions we are asking ourselves @least.

“It means that we are getting popular”

Well if this is what it means to be popular, I am not sure its a good thing.

Yesterday we came under attack from a DDOS attack…There were 10 IPs involved in the attack and they were trying to flood our servers to bring them down. We detected the attack quickly and tuned our firewall to block the attack…it took a while as we were trying to get the IPs etc. of the attackers from the same machine that was under attack.

We take the quality of our service very seriously and this was the first time we were down in months. I want to assure you that we are talking all the steps we can to ensure that such attacks do not happen again and if they do happen, they have no effect on the service. Following are the things we are doing:

1. Tracking down the attackers

2. Improving our Firewall service with a number of additional automated rules

3. Laying out our applications in a distributed setup so that we are able to communicate with the users even when we are under attack

Anything else we should do? Let us know?

I was having a swim in the community pool when Jitendra called me on my cell (I stepped out and picked up the call when the ringing became incessant).

Jitendra: Hey, Indus, where are you?
Me: Having a swim
Jitendra: Dude, sezwho.com seems to be down — There are tonnes of alerts, have you seen them?
Me: Well, I was at Midas, getting my brakes done and then jumped in the pool as soon as I reached home.

The heat of the issue dried up the swim pants and I was at home in no time. To our excitement, there were a few hosts continuously pounding our blog; making the rest of our services unavailable. On further investigation we found that we were under a SYN flood. Unfortunately, our blog is hosted on the same server where other webservices live. So, pretty much the core platform was unreachable for 3 hours or so. However, the image servers, databases, etc. were OK as they reside elsewhere.

We did a few things with our blog, watched the logs, but it continued as our current firewall was not able to block the SYN flood. Hence, we tried something simple by banning the IPs one at a time. Finally, we got the issue under control (for the time being). Here’s the graph from our server monitor.

Here’s our ToDo:
1. Upgrade the firewall
2. Move the blog elsewhere on a different host (and probably a different sub domain as well)
3. Monitor network -DPARANOIA

One of our creative users, Turnip, has created a custom set of stars (actually they are turnips, rather than stars ) for rating UI. He even did custom processing stage animated gif…Really cool and he thinks it was easy (we are looking to make it even easier soon)…See below (check it out by clicking on the image) :

We love this kind of customizations and really want to make sure that we provide you the full flexibility to enable these things…If you have done anything interesting like this, let me know and I’ll highlight it here.