I had a great support with me on the course — my wife & daughter who were waiting on agreed spots at km 10 and 32 to give me 2 and 2 gels so I did not have to carry them around. I have to say I really hated every single one I took that day and decided to skip the last. They probably gave me some energy that I needed during the race but it tastes so horrible! There must be something that will work better so I will try to experiment some more. During my training runs I love to eat almonds, nuts and baby bell cheese :)

Before the race started, I found two pacers for 4.15 which was my goal and learned how they plan to run. I somehow felt that it would not be wise to run faster than 6min / km at the beginning which was pretty much what they were planning to keep during the race. It was quite hard to keep the pace this slow the first several kilometers. I felt like I must be running much slower than that, probably due to excitement and the atmosphere. But it probably paid of in a great way. Basically the first 20 km were super easy and didn’t feel like running at all. 20 — 30 was still very good and last 12 was a bit harder — my legs were getting stiff. In the first half I managed to save some 6 minutes against the plan so I thought it woud be cool to make it under 4.10 and tried to keep the pace under 6min/km since that.

I do not have pictures to show but there were some crazy dressed people in the race. Japanese girls with 6 layers of clothes and huge rabbit years, guys dressed like Spiderman completely covered up including their head or some guys with a really huge backpacks. For those who haven’t been around, it was a pretty hot day around 22 Celsius and lot of sun. Since I had such a bad result on the half marathon just a month ago because of overheating I didn’t want to face the same issues. The race organizers did really a great job and practically every 2,5 km there was water available (and every 5 some energy drinks, bananas and oranges — yummy!). I took a sponge on the first refresh station and always poured a lot of water on my next and back and kept the wet sponge on the back side of my neck all the time. I guess this saved me since I felt super cool the whole race although it was warmer than a month ago at the half marathon.

Last 200 meters was the moment to sprint a bit so everybody sees how great I feel :) And in the finish was already waiting my support team to give me hugs and kisses. There was some pain during the last kilometers but already an hour later, while eating with my friends who were running that day too I was already thinking about next marathons to run! Also, I bought new shoes the day before — Saucony Men’s Grid Type A4. They were only 180 grams and felt so good that I decided I have to use them the next day. Pretty stupid idea, right? Buy new shoes and go run maraton in them the next day but it worked out well and they probably saved some of my energy too.

Overall, it was a great day. Since I signed up for half marathon at Silva Nortica run already few weeks ago I immediately sent them an e-mail if they can change my couse from half marathon to full marathon. Already during the night I thought it wasn’t the best decision since the pain came and it hurt everywhere from my ass and below. Since bones probably can’t hurt I was quite surprised where everywhere there are muscles on the human body… But that eventually went away :)

So the next race is Silva Nortica in 3 weeks and it will be my first cross marathon. What doesn’t kill you makes you stronger!

At the end, I would like to thank my friend Jaromír Vicari who showed me the beauty of running and went out with me to run countless times.

As a startup working with Linux, we were logically attracting other Linux people (both developers and admins) so our competency in that field kept growing and even if customers asked about it every now and on, we couldn’t deliver anything on Windows. We didn’t have Windows knowledge and we didn’t know how to support it (you can do a lot of things by clicking, right, but we didn’t have any idea what to do when the server is overloaded, crashing or throwing unknown errors).

Cisco UC Manager (previously Call Manager) was always a Windows application. It was back in 2002 and when we heard about that for the first time, we were really laughing to the idea of building a PBX (that is supposed to run no matter what) on top of Windows. It proved to work over time, also because practically any installation was sold as a cluster of minimum 2 servers and redundancy worked really well. One of the Cisco’s arguments at that time was that PBX now becomes part of your IT. It will not be anymore this black box that only few people understand. All you will need is a web browser and that is how you configure even the most complex settings. Sounds pretty compelling, right?

Our customer #3 was Czech Insurance (Česká Pojišťovna), local market leader in insurance business. They had some kind of policy that every Windows server needed to be connected to a monitoring platform a special client for that purpose had to be installed on the server. And here came also the first bigger problem. UC Manager did run on Windows 2000 Server but you were not really allowed to install there any other SW otherwise the installation would not be supported by Cisco.

So here is a story, how your PBX runs on standard Windows Server but you cannot really install and configure there what you want, only what the vendor allows (which is nothing). And it is fully understandable. If you would install there some application that would reduce the performance, negatively influence some installed libraries, maybe require other Java version runtime it would be real hell for Cisco to support it.

Our approach was a bit different. We were supporting CallREC on one Linux distribution – CentOS. What really helps is that CentOS comes out of RedHat, only there is no official support from the vendor.  And RedHat is a know name. We always had the installation of the OS and CallREC together and you didn’t have to make any choices about which OS packages to install and which services to enable. It was our problem, to make sure that just enough is running to make sure CallREC works.

Then Cisco released their UC Manager version 5 on Linux. It was an appliance approach since the installation of UC Manager and the OS was one and the same thing, you couldn’t separate it. And the OS is CentOS as well (I believe they call it Cisco Unified Communications Operating System).  The story was at that time that now there is only Linux version and in the future there will available both – Windows and Linux. Then came version 6, Linux only again and when version 7 came out last year it was clear there will be no more Windows version.

And it was a good decision from Cisco. If you have an application that you can fully manage from web-based GUI including all the system parameters, why do you care what what OS it runs on? You are not allowed to install to the same HW other applications anyway and your admin can find another hobby then “tune the OS”.

I read a very interesting blog post, which is dated back to summer 2007 from Srinivas Krishnamurti, Director of Product Management and Market Development at VMware called Get Juiced! Srinivas is describing the concept of JeOS (Just Enough OS) where the operating system is super tailored to the application. So instead of multipurpose, large and open OS the vendor “packages” the OS together with the application tuned up for maximum performance and security.

Cisco helped us in a great way. ZOOM always had some challenges explaining its customers why our call recording runs on Linux only. It was for them a black box, they didn’t have Linux admins and sometimes were saying “we are a pure MS shop”. Then Cisco came with CUCM 5 and we could say “CallREC is running the same OS as Cisco UC Manager, they actually chose the same distribution!” Suddenly, customers were OK with Linux and since about 2 years ago it is never a topic.

And it has benefit for everybody. We do not spend redundant time on testing our product on 2 very different platforms and can invest the manpower into new features, easier administration and implementing some OS specific enhancements that make our application on “our” operating system super fast. And now we have CallREC as a VMware Virtual Appliance too so it’s really easy to download it and give it a test ride.

I was actually thinking about some screenshots to illustrate that but I do not want to insult anyone :o)

Would you trust such recommendations? Do people really think they have any value? But still, I see such scenarios almost daily. I still do not understand why people recommend someone just because the person asked them to do so. You can always ignore the request or politely decline. What happens is that you logically adjust your respect to a person that recommends someone you know and about whom you think is a schmuck.

Recently I got contacted by a lady I once met about 8 years ago in a pub, asking me about recommendation. I politely refused since even if I know her, I actually never worked with her. She replied that it is OK, that she understands, but she is trying to get some recommendations to her profile since nobody wants to invite her for interview without references.

Really? People actually decide about inviting a person for interview based on the amount of LinkedIn references? That is scary! Sometimes I also get CVs with proud statements in the intro letter like “I have 14 recommendations on LinkedIn!”. Who cares about the quantity if quality is questionable?

It’s like search engines before Google came. It was enough to have a particular word many times repeated on the page and even more times repeated in the keywords section to be displayed on top of search results. And then came the PageRank and suddenly you could really find relevant stuff on the web.

In order to make a meaning of the references, LinkedIn and other business networks need to think about some kind of “TrustRank” related to people. It could be a formula, calculated (pretty much like the PageRank) from different values from your network.

Here are some ideas about what the formula could contain:

• Probably it has to work against the principle of PageRank – the fact that someone is many times recommended doesn’t increase his/her credibility – it can just mean he/she annoys a lot of people to get recommended.
• Is should dramatically lower the credibility of people who “cross recommend” each other within a short period of time (a few hours / days?).
• It could capture people’s credibility, real achievements and reputation from other sources on the web and blend that in. I know a few people who rarely recommend anyone. And if they do, I tend to really trust them.
• If a person is recommended by a colleague or boss you could track how long they were working together before the recommendation was created.
• Recommendations closer to the end of someones job position should be also discounted since people tend to pro-actively ask for recommendations when looking for a new job (I’m not saying that is bad, just that recommendations might not have that high value as opposed to a spontaneous recommendation when someone impresses you with their job).
• The users in the network could (maybe anonymously) score and comment recommendations of other people in their network to share if they agree or not.

So far, I recommended only 2 people, both colleagues from ZOOM that I know for 5 and 8 years on respectively. You have only one reputation on the web, do you? And even deleted posts still shows in the results…

Happy to hear your comments.

At that time I was a college student and building computers from individual parts as a job to make money. It went very well actually since there wasn’t really any chance to buy a PC for a reasonable price on the shell — forget Best Buy or similar. Since more than half a day I was at school and I didn’t even dared to turn the phone on during brakes I needed voicemail to hear if there was any customer calling me for business.

Very few people had cell phone at that time, it was not possible to send SMS between the 2 different operators and web SMS portals did not exist for the next many years. So voicemail worked well.

When I fast-track 10 years to the future, even ZOOM made its own product — VoiceBOX — based on the fact that people want voicemail (we discontinued it about a year ago, putting all our energy in call recording and QM solutions) . At that time Cisco’s voicemail called Unity was just crazy expensive (well it contained some features like speech recognition that nobody wanted to use) and we saw a gap on the market. So we did a simple IVR script that did the prompt playback and message recording and build on top of the web-based GUI to manage users, groups, replay messages and even a little workflow when one mailbox has been shared with more users — like support number in afterhours.

If I look back, apart for maybe the first few years, I never liked voicemail. Visual voicemail (which we also implemented for VoiceBOX on Cisco IP phones XML interface) is better than the horrible IVR retrieving but still.. In Europe most people hate voicemail and I always hated when somebody left me a message saying “Hi, it’s me, call me back when you can”. If I see a missed call on my cell phone, I call back when available unless I really do not want to speak with you so leaving me a voicemail wouldn’t help. And the only people calling from hidden number are UK call centers offering me sure financial investment, better insurance or an amazing conference for my company with guaranteed 34 face to face meetings with CIOs of Fortune 100. No thanks.

About last 3 years I always used to alter my welcome message in voicemail before holidays. Something like “Till so and so I’m away from my phone, I do not read e-mails and call my colleagues for sales or support”. I always recorded this message at the airport some 10 minutes before I boarded the plane, then forgot after my return and then during the next x months got many mails saying I should update my welcome message.

So here I go, decided to bury my voicemail for good. I assume if someone has my mobile number and has an urgent message, SMS does the job very well or an e-mail message thanks to the great BlackBerry that my wife doesn’t really like. So I have two types of holidays now — with and without BlackBerry and I tend to enjoy the later much more.

BTW, the voicemail transcription service that Google Voice has looks really cool, that would probably make me stay. But still, SMS, Skype, IM, e-mail or BlackBerry chat where you actually see not only that the message was delivered but also when it was read make the job for me.

It was an interesting conference. Hosted in Microsoft’s Conference Facility in Mountain View, I believe there could be close to 200 people, almost all entrepreneurs. Microsoft has really cool room there with a power socket and Ethernet in every seat. Guy wanted to sell us patch cables for 50 USD a piece but his extra source of revenue has been ruined by free WiFi access we got.

There were some great speakers including Chris Anderson, Editor in Chief of Wired magazine and author of FREE: The Future of Radical Price, Paul Graham, partner of Y Combinator, Scot Hintz, co-founder of TripIt and many more. I expected to feel like Alice in Wonderland and often I really did. The whole Web 2.0 and the new way of making money on-line was the core topic to learn more about.

Unlike most other conferences, good 75% of the attendees where online by one or another way. 50% had their laptops open (some 50% Macs I would guess), others had smart phones or tweeting by texting. A #revenuebootcamp hash tag has been used during the whole conference and you could see projected on the wall during the whole time people’s comments with 10 sec refresh.

Basically teens are not open to pay for anything :)

It started pretty funny. In an opening panel discussion, there were 4 teenagers out of the 5 panelist (plus Guy as the moderator) and the long story short is that they are practically not open to pay for anything. They do not click on ads, they are not open to pay for TV, if most of the services they use would start to charge, they would leave. Couple of observations: they all use Gmail and love it, eventually would be open to pay a little bit for the service. Most of them use Facebook, but half of them would leave if they would have to pay. BTW, I do not think they have to worry about Facebook started to charge its users for using the site. They get so much valuable info about its users so they can sell very targeted ads. They also like XBox Live, which they find kind of cool. That’s it, everything else on the Internet should be free in their eyes. The future generation of our customers is not open to pay for anything.

In the next panel, we focused on how to get some traffic to your site. The single most important output I have from this discussion is that you should really focus on quality content first to make sure that users come back and find your site useful. All the SEO and AdWords and other ways how to make users visit you for the first time are wasted if all they do is click the back button and will not return. For anybody that plans to provide an on-line service, we need to think about how to start making money from day one. There are some exceptions like Facebook, Twitter, Youtube. But that is it, everybody else needs to think about the revenue model from the very beginning, not hoping to get this huge critical mass so investors will pump the money in the business with hope to monetize it in the future.

Dion Lim, President & COO of Simply Hired mentioned that “SEO is not a project, it is a mindset”. It is not something you think about as one of the features, it’s just one of the foundations of any successful on-line site today. If you have money to play with, you can also experiment with colors. Agar Kamdar, Product Manager of Webmaster Central at Google mentioned, they increased a click through rate on Gmail sign-up button by 20% changing the color from green to more blue, pretty amazing. As I remember Amazon paid a lot of money for a research that showed that yellow is the right color of buttons for them.

Chris Anderson during keynote

Then Chris Anderson had a keynote, basically about his new book and new phenomenon – makings things free and making money somewhere else. The book is very inspiring, I’m in about a middle now.

Chris Anderson signing his book

The next panel was about the future of advertising and if this model is dead or not. It surely isn’t. Every business needs to find the best revenue model but advertising powered content still works pretty well. Tim Kendall, Director of Monetization at Facebook mentioned that social networks can provide a platform to even better serve relevant ads to its users – if you have same hobby as your fiend and he clicks on some, there is a higher chance you click on the similar add too. This can mean much higher percentage of click-through rates, higher content relevance. Imagine what Google will know about you with it’s Google Voice offering.

The next panel – Beyond CPM, Innovative Models for Generating Revenues – was really cool. We learned that IMVU is making millions of $ by selling virtual clothes to the avatar characters in on-line chat. And they do not even make them! Third parties make clothes and accessories and do revenue sharing with the site. I still find it pretty amazing, how much is spent on virtual stuff. They also introduced a virtual wrap paper, started at just 2 cents and increased the price all the way up to 2 $ and users were still buying it – just a background in the text field when you are sending someone a gift (that you paid for as well).

Jeremy Schoemaker, Founder of ShoeMoney Media Group was really funny. Telling his story how he got his first check from Google AdSense for 140k $ for just a month (!!) and he makes money now blogging and tweeting (next to offering his services in exactly this area). He got from “an unnamed” company that launched it’s new search engine about a month ago 400 $ for every tweet where he mentioned that (limited to one tweet per 4 hours for duration of one week). Pretty cool, right? He has 38k followers. It was a eye opener for Guy – he said he did tweet too about this but for free :)

Fireside Chat of Guy Kawasaki with Mike Moritz and Paul Graham

Last session was Fireside Chat of Guy Kawasaki with Mike Moritz (Sequoia Capital) and Paul Graham (Y Combinator). Paul invests some 20k $ in start-ups and tries to make them up and running for the first 90 days so they are ready for next rounds and expansion. I read an interview with Paul in the latest issue of Inc. about the same topic. Mike is a funny guy too. During the chat he obviously felt very comfortable, taking of his shoes while playing with them with his legs showing us very nice blue socks with white dots :) His message was that they search for passionate entrepreneurs who, 7 years from the investment could say that choosing Sequoia was the single best decision of their life.

So, pretty packed day! I exchanged business cards with couple of people there, some at the stage of “we are two guys working on this next big thing”, which is so inspiring to hear and reminds me when all the people at ZOOM fit into one small office and we were all running at 11.55pm out to catch the last metro to bring us home (if we didn’t plan to sleep there that night).

Couple of last thought and links I learned:

• A lot of other ideas can be found from many people tweeting during the conference. The hash tag is #revenuebootcamp which can be displayed easily using TweetChat
• Crazy Egg offers pretty cool service to see how people use your site and where they click even if it is not a button (heatmap from mouse moves and clicks and some other interesting stats)
• SEO Browser allows you too see how crawlers and search engines see your site
• seomoz.org is a comprehensive site for everybody who things and dreams of SEO with a lot of free tools and a pro edition

And because of a little jet leg I got up on Sunday 7am to write this post :)