My Biggest Career Mistake

Image via WikipediaIf you've been around for any length of time chances are good you've had a few significant bifurcation points in your career where you had to make a choice. Sometimes those choices work out well, sometimes they don't. Skill and hard work play a part in the outcome but luck does as well.

Around 1997 I discovered a then obscure and very new open source database known as MySQL. MySQL has grown up a lot since then. It's currently used by Facebook for instance and provides at least part of the back end database services for many other well known sites. At that point though it was just a promising new comer that had three key advantages over it's competition.

1. It was free
2. It was fast
3. It was multithreaded

Point number three was actually the clinching differentiator, at least for me. What "multithreaded" means is that it was capable of handling more than one request at a time. This made MySQL ideal for small to medium sized web sites that wanted some sort of active content. MySQL was optimized for read speed at that point in time and given the demands of the day it was a great solution. It lacked things like sub selects and other more advanced SQL features but it had enough functionality to get a lot of jobs done and the price was right.

When I got involved the documentation was a very thin. Monty Widenius and David Axmark were the two driving forces behind MySQL and at the time a lot more attention was (rightly) being focused on fixing bugs and adding features. This made perfect sense and it gave me an opportunity to step in and help. I wrote or rewrote the majority of the documentation for MySQL over the first couple of years of it's existence. I also wrote a couple of articles that appeared in the now defunct Web Techniques magazine that helped champion both MySQL and PHP. Eventually though I was faced with a choice.

I had an opportunity at work to step into a management role. This was a long term goal of mine so I took it. I was managing people who were older than me and I was in a bit over my head at the time so I had to disengage from the MySQL community and the work I'd done there. Several years later MySQL was sold to Sun Microsystems for a billion dollars.

I was never an employee of the company that owned MySQL though I did receive MySQL license number seven for free in thanks for the work I'd done.

I can't help but wonder what might have developed for me professionally if I'd passed on that management position and continued my involvement with MySQL. I learned a lot in the management job and had an opportunity to work with some great people so it wasn't a bad experience from that perspective. Weighing that against what might have been if I either hadn't been given the management opportunity or chosen differently does occasionally keep me up at night though.

Klout Vs. Peer Index (Again)

Image by Intersection Consulting via FlickrI've written about both Klout and Peer Network before. Both are social network influence rating sites that purport to give individuals and companies an understanding of who the key movers and shakers are in the social network world. The way they do this is by using arcane and possibly questionable math formulas to evaluate people and companies on line interactions and networks and assign a score. My main concern continues to be the fact that I don't believe it is valid to measure multiple sites and somehow generate a single value that is comparable across more than one person. In other words, I feel good about tracking my personal progress but I don't feel good about comparing my Klout score to somebody else's since I may have a total different model of engagement than that other person. By this I mean for example that somebody else might use Facebook and Twitter as public facing means off social networking while I use Facebook only to communicate with friends and family and instead use this BLOG and Twitter as my public facing persona. I'm not going to be sending out thirty or forty updates a day on Facebook because of my use model while a person who is using Facebook as a means of communicating with a wider audience probably will.

Setting all that aside for the moment I'm ready to declare a victor, and the race hasn't even been close. Peer Index has always been extremely slow to update. In fact there are portions of my profile that have never updated. I even opened a ticket with them a couple of weeks back and while I did eventually get an acknowledgement nothing has changed as of today.

In the meantime Klout has made regular updates. Like clockwork I can login  every day and see what if any progress I've made in raising my Klout score. I might question the science behind Klout's scoring but I don't question the apparent influence they already wield or their ability to execute on their vision.

If my experience is any indicator Peer Index is in a lot of trouble. They are slow to update, slow to respond and are steadily being out innovated and out performed by Klout. This isn't a good situation to be in.

I'm going to keep tracking both for awhile more but at some point I'm going to stop bothering with Peer Index if they don't start to turn things around quickly. They just aren't generating any value for me right now.

Related Articles:

• I Can't Deny That Klout Likes Me Right Now (emkey1.blogspot.com)
• Klout & Peer Index Followup/Empire Avenue (emkey1.blogspot.com)
• Klout & Peer Index (Social Network Influence) (emkey1.blogspot.com)
• Facebook + Twitter = Klout (blogs.forbes.com)

Cisco/Marketing/Borderless Networks

Image via WikipediaThe first time I heard the marketing term "Borderless Networks" the obvious semantic interpretation made no sense to me. To be clear, I'm talking about computer networking, not human networking though the former enables the later on the Internet.

Many years ago when the world was a much more innocent place nearly borderless networks existed as computer security was essentially non existent. There's a lot of history that I could explain here but suffice it to say that the ideals of the early Internet pioneers proved to be naive and over time we've seen  layer upon layer of security added to networks in an effort to keep unauthorized people from doing not so nice things with data such as credit card numbers and other sensitive stuff.

Marketing types in the tech field have a tough job. Ideally they need to craft messages, products and strategies that appeal to both the technical and the more business oriented people who make purchasing decisions. This is particularly true of big picture marketing concepts like "Borderless Networks" which encompass a number of different products and are meant to describe something revolutionary in a way that excites the imagination and frees up big wads of cash.

We technical types tend to be very literal minded. When we hear "Borderless Networks" We think of... well, networks without security trust boundaries; at which point we get some combination of nervous/dismissive. From a logical perspective a network without trust boundaries is a completely absurd concept in this day and age. It sounds like Cisco is suggesting we move into a very bad neighborhood and leave all our doors and windows open with signs that say "Rob me please!" posted everywhere. The thing about the Internet is that from a logical perspective everyone lives right next door to every other neighborhood including the worst one which is why we need and have trust boundaries AKA borders.

Suffice it to say that on the technical side I don't think "Borderless Networks" is a very successful marketing term.

Of course from a logical perspective Cisco isn't really trying to sell us networks without borders. Instead they are pushing the concept that network trust boundaries should not be dictated by physical locality or unique but easily faked attributes of the hardware you are using to access a particular network resource such as your IP address.

Right through today most network security has been focused on restricting access via some combination of the IP address of the computer a particular person is using and some form of password authentication. This is a very course way of handling access control and one that doesn't provide a lot of flexibility in terms of where people can work and what data they can access. Where we're going is towards more of a role based approach which takes into account who a person is and their role rather than where they are located..

So, if you're role in a company changes your access to resources will change just as soon as IT flips a few bits. In the case of Cisco this change would likely happen through some combination of changes in Active Directory and Cisco's ACS product.

In the current model such role changes require a lot more work and tend to be very haphazard. This means there is an inevitable transition period when people in new roles still have access to information that may no longer be relevant to their job and lack access to information that is very important to their new role. The old model creates both operational and security issues.

So, while technical types are apt to hear "No logical boundaries" what Cisco's marketing wants us to hear is "No physical boundaries" when the phrase "Borderless Networks" is used.

To make this role based magic happen requires big changes in the way computers, applications and network infrastructure interact. The technologies that enable this to happen are what Cisco is referring to when they say "Borderless Networks". 802.1X, VLAN, NAC, ACS and dynamic VLAN assignment are all part of this.

The ultimate goal here is to enable an increasingly mobile workforce to access corporate resources in a reliable and secure manner, and sell lots of hardware and software of course.

I'm not convinced that Cisco's marketing department chose the best possible labeling to express the concepts encompassed by "Borderless Networks" given their bifurcated audience but I understand the complexities they were facing. Trying to come up with a pithy and descriptive phrase that will appeal to both the technical and business crowd is a non trivial task. If you're going to have to settle for hitting a home run with just one or the other than the people in the suits are where you're going to focus your efforts.

Related Articles:

• Network Defenses - Security Through Network Design (codingexplorer.wordpress.com)

The Twitter Bots Have Found Me

I've been gaining some additional followers on Twitter recently which is always nice. Some of them are clearly not real people though. I realized this when I got three new followers in a very short period of time. All had odd names and exactly the same tweets sent. Not surprisingly the tweets were advertising stuff. I suppose this a right of passage of a sorts but I'd just as soon not have to go though it.

There has to be an economic incentive for spam or there wouldn't be anywhere near as much of it as we see. I used to wonder who was falling for these obviously shady deals. Then a relative developed Alzheimer's and it became clear to me where at least one source of sales originate. When you're mental faculties start to decline the offers in spam look a lot more appealing. Realizing where at least some of the "customers" are coming from made me dislike spam a whole lot more than I already did. 

In addition to spambots that send tweets I've also encountered @emkey1tweets tweets that are links to spam. None of this is really news if you've been a heavy Twitter user for any length of time though. 

One secondary effect of all this spamming is that it makes social network influence harder to measure since those scores depend in part on @ activity and followers. The scoring includes determining the quality of an individual or companies network so as long as the algorithms in use take this problem into consideration it's hopefully not a significant issue. I can imagine some people "courting" spam bots to up their scores.

The stability on Twitter recently has been poor. Maybe the spam bots are to blame but there seem to be a lot of rumors going on about internal strife and general discontent. Likely a big part of the problem is that Twitter has been much better at generating revenue for others than itself. At some point investors start to want to see some sort of return on their money and employees start to wonder if they are ever going to get to cash in on their hard work. Like it or not a big part of the economic and intellectual engine that drives the valley is based on the assumption that hard work and success will be rewarded with large wads of cash. Twitter's endgame is far from obvious at this point and they may be starting to suffer because of that.

In spite of all the recent hiccups and other issues with Twitter there are an amazing number of very interesting and cool people available to follow if you spend a little time looking around. For instance, enter "Entrepreneur" in the search box and start working your way down the list of names. You might even be able to start a conversation with somebody who is highly knowledgeable in a field of endeavor you care about. 

Image by GE Healthcare via Flickr

Related Articles:

• Things You Should Know About 'Twitter Jail' (techpluto.com)
• Twitter Spam (talkaboutlocal.org.uk)

Borders/Apple/Google+ & Facebook

Image via WikipediaYou shouldn't notice any differences on the reading side of things but I've switched over to the new Google blogger beta composing tool. So far so good, though there have been a few bumps in the road. I had to do some manual editing in raw HTML mode to correct a couple of formatting anomolies.

The new composition tools don't seem to offer much additional functionality but it's early going yet. I decided to be proactive in making the change but I don't recommend that to everyone. The interface is a fair amount different and there certainly is a learning curve. It's probably a good idea to wait until it's a bit more polished before making the jump.

Borders
The final dissolution of Borders wasn't a complete surprise. When they closed a bunch of stores earlier this year the writing was on the wall. My wife and I were frequent visitors to our local Borders over the years and it was kind of  a surprise when it was one of the first ones to close down. The store  seemed to be busy and the staff were friendly and attentive. It was hard for me to understand how they couldn't be profitable given what I'd observed.

One thing that did bug me was the way they did their coupons for the Borders reward program. On the plus side they emailed the coupons regularly and the offers were often attractive. On the downside initially you had to print out the coupon and bring it in. They would then take the coupon. If both my wife and I wanted to use a particular coupon we had to print out two. At first I thought there might be different codes for tracking but when I compared two of the coupons we printer out they looked identical.

The Borders rewards program included a card. All the coupons were clearly in their point of sale system. All the coupons for a particular offer were so far as I could tell identical. Why then did they insist on a printed copy?

Later they did change the rule so you could show them a copy of the coupon via your smartphone but the whole thing was annoying and inconvenient which was kind of counter productive.

Maybe there was some obscure reason for this bizarre rule but it's certainly a head scratcher and it made me start to think that the problems Borders was having were rooted very deeply in the corporate side of things.

Barnes & Noble has done OK using essentially the same business model as Border's so you can't explain this bankruptcy by simply blaming Amazon, EBooks and other technological shifts in the book industry.

Borders has now joined Circuit City, CompUSA and many other large chains in the march down liquidation road.

Apple
I don't actually have anything to say about Apple other than WOW. Seriously, what they are doing in this economy is just amazing. Mostly I mentioned them because putting Apple in the title of my posts is always good for a few more page views.

Imagine a smiley here if you like them, pretend I never brought up the topic if you don't.

Google+/Facebook
I'm expecting to start hearing about G+'s growth slowing over the next month or so. It may just be me, but I'm seeing pretty much zero uptake by my less technical friends and contacts. G+ has become the geek choice hands down but that isn't going to carry them beyond a niche player and being a niche player in this space isn't very exciting as I've noted before.

In some cases people are apparently worried about the potential privacy issues of getting a Google account. That kind of confuses me. Both Facebook and Google know an amazing amount of stuff about me. They are both large companies with many fallible human being working for them.

Personally I prefer Google's approach to privacy right now but I may not ten years down the road if their management changes or undergoes a change of heart. Anyone that truly cares about their privacy should probably think very hard about the folly of putting a tremendous amount of personal information in the hands of any big online company. Before doing that though you have to ask yourself what kind of life you want to lead because like it or not the 21st century as we currently know it here in the US and much of the world is driven in large part by the ready flow of advertiser friendly personal information.

We collectively chose this path when we decided we wanted our Internet to be "Free". Employees and share holders need to be paid and so advertising became the economic engine of the Internet. When a business model is built on advertising you've given the companies in question a heavy motivation to optimize their means of bringing money in. In the case of advertising optimization means developing very sophisticated methods of tracking and categorizing people and this in turn leads to a significant reduction in privacy.

At some point in the not too distant future G+ needs to start making inroads to the mainstream. I'm personally hopeful that they will, but I'm starting to have serious doubts.

Related Articles:

• Privacy watchdog sets sights on Google (canada.com)
• Google's Biggest Problem: It's Too Rational (pcworld.com)

Volume/Text/Audio/Video

Image via Wikipedia

When I wrote yesterday about the efficiency of text in communicating ideas and concepts my primary motivation was to point out that no matter what medium is used, everything boils down to ideas that can be most efficiently represented by a finite number of words that take up much less space than video, images or audio. I think this is one of the reasons that audio blogging never took off the way many people expected it to. Audio BLOG's are difficult to search through and the information density just isn't as high as good old text.

Video arguably has taken seen more success but I'm not convinced the level is as impressive as some would claim. It's true that Videos very low information to byte density makes it an increasingly large percentage of total network utilization but one factoid I heard at Cisco Live 2011 put the amount of video uploaded to YouTube per minute at forty eight hours. That really underwhelms me. How many BLOG postings do we see in an average minute? I don't have any stats but I'm betting it would take a lot more than forty eight hours to read one minute of BLOG posting output.

I'm not trying to convince anyone that audio or video are bad. I've found a lot of value in both when they are done well. What I am pointing out is that in spite of all the hype, particularly for video; text is still the most efficient means of expressing ideas and concepts.

Video draws us in though as we are programmed by many years of our ancestors living out in the wild to constantly be looking for and assessing motion. When we go to a web site and a video starts playing somewhere on our screen we can't help but look at it; and that is what advertisers count on and a big part of the reason for videos popularity. That small video ad is going to be 99.99% of the total bandwidth of just about any web page in spite of the fact that it will in most cases be a tiny fraction of the total information delivered. From a revenue perspective though that ratio is totally flipped. Video advertising is a big chunk of the revenue pie for any site that employs it.

Moving pictures have entranced us since their invention more than a century ago and that isn't going to change. Stories told with words, whether written or spoken have entranced us for a lot longer though and I think it would be a mistake to forget that fact.

I'm not really a Luddite, but I do occasionally play one on the Internet.

Related Articles:

• The Voynich Manuscript: will we ever be able to read this book? (telegraph.co.uk)

Plain Text/Information Density

I've been hearing a lot of talk recently about how Video will take up the vast majority of the network bandwidth soon. This is great if you are a network technology vendor but I wonder how much value is being gained in relation to the increased cost?

Prose text is an incredibly information dense way of sharing ideas and concepts. Tolstoy's "War and Peace" had 560,000 words. Although it was originally written in Russian and French lets multiply that by the average number of letters in an English word which is 4.5. We end up with 2.5 million characters, or approximately 2.5 megabytes. That's not enough to store a single twelve megapixel JPG image of good quality. The dichotomy becomes more pronounced as we move to less verbose classics such as "To Kill A Mockingbird" or "The Great Gatsby".

I'm assuming 8 bit ASCII encoding here but even uni code only doubles the size so for the sake of simplicity I went with ASCII.

Audio is better than video or images from an information density perspective but you'll still only get a minute or so of good quality MP3 stereo audio for that 2.5 megabytes.

I'm not arguing against the use of audio or Video on the Internet. When done well either medium can be a great means of communication. The costs associated with poorly done video in particular though are gargantuan when compared with text. Plain old text allows us to render ideas and concepts in a form that is phenomenally efficient space wise. As an added bonus text can be compressed without loss quickly and easily to 20% or less of its size.

From a business perspective there is no greater bang for the buck than prose text when we're talking about the network.

The incredible increases in data being generated each year are often cited by industry leaders as some sort of wonderful magical thing with the implication that the amount of information being created has some sort of huge positive impact on the pace of innovation. The truth of the matter is that while we're creating vastly larger volumes of information each year the actual information density is dropping nearly as quickly so the net result isn't as impressive it appears at first glance.

All that hype does help sell a lot of hardware though.

Image via Wikipedia

Related Articles:

• How Bits and Bytes Work (mgitecetech.wordpress.com)
• Five Awesome ASCII Art Generators (theotakuologist.com)