I didn’t make it to mesh or meshU this year. Long story. While I missed out on a lot of great talks I’m able to find a fair bit of content elsewhere. First, MeshU.

Leigh Honeywell of hacklab.to has her slides on writing more secure code up on her blog and I’m sure she’ll chat with you about security at freenode’s #hacklabto or at the lab itself.

Ilya Grigorik has a great content packed blog post including his slides on his event driven architecture presentation.

Carl Mercier has his piece on building and then selling his startup available in video and with slides.

And that’s just MeshU. I’ve barely dug into the mesh conference itself, which is available as an iTunes video podcast and through the meshTV mDialog page.

I’m halfway through watching the segment on the future of news. Spot.us sounds very cool as a means to crowdsource funding for articles written by a journalist, and potentially paying the crowd back through licensing.

Daniel Debow of Rypple is speaking. Talking about startups.

First startup was Workbrain. Grew quickly since 2000, sold in 2007 for $227 million. Grew revenue by about $20 million US per year since 2001. Latest startup is Rypple, started about six months ago.

What type of startup? What are types? Usually brings up images of Zuckerberg and Facebook, Bill Gates and Microsoft, RIM.

Do it yourself. Do it in a garage. Or join a friend, find an experienced partner, work on small projects for local startups.

Join an early stage, fast growing company. (Ed: Do you join before or after funding? I misheard that part. I suspect after funding judging by bit on risk later on.) Small sampling: Learnhub, gigpark, freshbooks. Why? You’ll have a lot of impact. Every day in a startup is like a quarter in an established company. You have autonomy. Responsibility. Fast paced.

But aren’t startups risky? Not as much as you’d think. Startup does not mean work for free. What did you think was risky? Big companies like Microsoft, Lehman Brothers, AT&T, Nortel Networks laying off thousands of employees. There is no such thing as job security any more. And what about risk of a job where you’re not passionate about? That doesn’t let you be creative? That doesn’t let you have an impact? What’s the real risk in a startup? Not starvation, or homelessness, or becoming a social pariah. It will be valuable experience. You don’t have a family to feed (yet.) Don’t worry about trying and failing. The real risk for a startup: not trying. You don’t regret failures as much as not taking an opportunity.

Find the startups. Take it as seriously as coursework. Search online. Ceck out barcamp, democamp, startupindex, startupnorth, crunchbase. Maybe you don’t bring up your startup idea but you mingle. Don’t be limited by geography. Volunteer informally on small projects. Don’t be shy, at worst they say no.

To get the job, be yourself. Be excellent. Show you’re useful and they’ll find something for you. Get the interview basics right! Details, details, details! Work with other people in school, sports, bands, charities to show you play nice with others.

PS: Try out Rypple. Rather than copy out notes about what Rypple is, I refer you to their site and the killerstartups.com article on it.

Related articles by Zemanta

• Some Tips for Startup Employees (killerstartups.com)
• Rypple.com – An Interactive Tool For Self-realization (killerstartups.com)

The speaker was Eugene Roman from Open Text.

About 10% of the crowd has heard of Open Text. Eugene’s job is to change that. Started at Waterloo from two professors. First project was to put the Oxford dictionary online. Made the first internet search engine. Largest ECM vendor as of 2007.

We live at a critical time in the tech business. We don’t have to worry about wireless devices, that’s been done. Laptops are there. Ubiquitous internet connectivity is there, but expensive. What we need to worry about is content.

What does Open Text do? They connect many platforms. Oracle, SAP, Office, Siebel Oracle, Exchange Server, etcetera. How to manage all this information.

Enough about the company. Now we’re welcomed to the digital world. Dr. Donald Chrisholm said that in 1979. Was given half a billion dollars to build the first digital phone switch. We’re in a world of instant communication, a world where information is power and is frictionless. A boundaryless global village. A world of highly useful, amazing and at times useless content. Mr. Roman’s gift: the glocal effect. Global and local at the same time. Immerse yourself in a region’s local content from anywhere. If you don’t understand that, you can’t understand the future of connectivity.

How can information be useless? Spam, SMSes from your ex. Information is like cholesterol, both good and bad. The world isn’t simply a digital world anymore, it’s a digital content world. Ever thought about what is the root world of technology? Greek word techne. It’s the Greek word for art. The great artists of the times were considered great technologists. Sometimes we forget this, but now content is coming back among technologists.

Digital disruptors. Easy prediction: everything that can be digital, will be digital. Yes or no? Yes.

What’s fun got to do with making money? Infotainment, entertainment, it’s all content.

We see appliances that create content, and content that creates appliances. Do you want to be in the content or the appliance business? Generally companies that are good at appliances are bad at content, but they’re great at presenting it to you. The actual success of the iPod is not the device itself but the ecosystem for the content. Devices lead you to deal with warranties, content is fire and forget. Produce once and replicate infinately.

Books are becoming obsolete. Digital can help you be green. Content today is dominated by GEMAYA: Google, eBay, Microsoft, Amazon, Yahoo, AOL/Time Warner. These are disrupting traditional communications and entertainment. Pick your content providers carefully. $13 billion in advertising revenue in the top 4 portals, $6.3 billion for Google’s US online advertising revenue in 2007. All of this is from vapour, in the content world.

Mixed media becomes rich interactional content. Did Gutenberg understand this? Have we thought about 2020? Students usually only think ahead to the next beer. But even if we tried, it’s so far forward it’s hard to think about.
We’re disadvantaged for this digital disruption because we have been bathed in bits. We have to learn from the time before to help us take advantage of the time that comes. First internet, don’t think DARPAnet, think morse code lines. First messaging infrastructure in the world, it moved the world from month-long international messages to seconds. Book to read: The Victorian Internet. Takes about two hours to read. It will change your life. Caused Eugene to think “everything that can be digital will be digital.” We live in the moment, with all our devices we have a mind block, unable to imagine life without our devices. Thinking about the Victorian internet will help us think of how the world will change in the upcoming digital evolution.

We are the millenial prosumers. We always have a computing device on us. Twenty eight years ago this was unthinkable. Computing is now a commodity. Getting into hardware now is the wrong end of the curve, it’s explosive growth is over, content’s is about to come up. How did Gutenberg understand rich interactional content? Ever seen a Gutenberg bible? Here’s what Eugene saw. Printed pages, also the effect called “illumination.” After they printed it they sent it to an artist to highlight and colour letters in major paragraphs with colour lettergrams and pictograms. An “illuminator” used gold ink, red ink, etcetera. The illuminators understood the human condition and how to get attention. Humans live on interesting things.

In the past, creators were few and readers were many. Now creators are many and are close to one to one with readers. “Copies and Seconds” is the next book we ought to read. The Xeography machine (photocopier) was the next age after Gutenberg. He figured it out in the year 1959. He made $150 million the year after he completed his invention of 19 years, and then donated it to American universities. “Chester never needed very much. He was all about giving back, especially to students. They are the future.” Why would he invent Xerography? He understood that many words would matter. As a patent clerk he used to hand reproduce things. Then the reproducable picture came along from Kodak. Digization is the link. The next link, Rochester in New York. Chester was from Rochester, Kodak started there as well.

There was a mixing of blood between Xerox and Kodak. There’s something about people that learn from eachother. Make sure that your company works in a good ecosystem.

Kodak is an excellent case study in a company that could have owned the digital age but did not. It was hard for their business model to change from silver on plastic, which they could charge a great deal for, to bits, which are free.

Massively reproducable interactable rich content is worth billions of digital bits (priceless.)

Content is evolving with bandwidht and personalization but there are no rules. No governing body of the internet beyond commercial interests and loose associations. This is good and bad. Alephs will change your world. Memexes. In this world of search and knowledge, we did not predict the rise of Google. People who studied memes could see that Yahoo would have been superceded by Google, though.

What makes Google, Google? Graph theory! Sergei Brin learned it from the University of Moscow, the best computation school. They had to think differently because they did not have much in the way of computing power.
Web 2.0 is extremely rich media disruption. Everything that is digital can be captured, clicked, used. Metadata matters. Clicks matter. Everything matters. Everything that is digital can be tagged. Unfortunately, e-mail is still the killer app. 75% of all digital knowledge is in e-mail. One day of e-mail is equivalent to ten world wide webs. The Blackberry is an e-mail appliance. Now they need to make the transition from e-mail to rich interactional content. They need to attach with a content provider.

How did Apple become Apple? Read “I, Woz.” They understand the Wozniak effect. It’s the only story of a true inventor in the modern era. What does ipod stand for? Internet protocol on demand, according to Eugene. They took the power of the internet and they made it highly useful in your pocket. E-mail still matters and will continue to matter. SMS will continue to matter. The problem with e-mail is liability. An e-mail that a company sends can get a company into a lot of liability. One of the products Open Text makes is the ability to back up and index e-mail. Particularly useful when seven years from now you get e-mail subpoenaed. If you don’t produce the ancient e-mail, an executive gets a contempt of court charge.

What did Carver Mead of Caltech say? Listen to the technology and find out what it is telling us. One real key: high fidelity. Think about it. Why is it important? CDs were cheap. Sony created minidiscs. MP3 players came out. The problem with MP3 is that it’s a lossy compression. Apple still managed to pull knowledge out of Dolby to make the iPod high fidelity. The quality of ipod earbuds on an ipod were roughly twice as accurate than anything else on the market with base earbuds or headphones.

Eugene’s quote to us: The global digita village is about extremely rich, highly interactional content.

Related articles by Zemanta

• Open Text Introduces New Versions of Desktop, Thin Client Content Viewer Solutions (newswire.ca)
• Open Text Extends eDiscovery Offering (newswire.ca)

In place of a day three report I offer a link to Ben Kane’s summary of the conference. It turns out that not only did Richard Stallman give the same talk as last year, he gave the same talk at the University of Toronto a few days later, and still hasn’t bothered to look up any of the figures that he prefixes with “and I think that..” You may as well just read Joey deVilla’s notes. While you’re at it, read how as a Microsoft evangelist he subsidized the Free Software Foundation. I guess the hundred bucks was worth the thousands of dollars in publicity. Clever move.

I started out the day with James Golick‘s tutorial on dynamic languages. Unfortunately he couldn’t get the projector working due to issues with his MacBook. Since it was a tutorial the projector was essential, so after a several attempts it was cancelled and people shuffled to the Facebook tutorial. In my web programming course I had already learned about the API so I took the opportunity to get lunch. I spent too much time out there and didn’t return in time for the next session.

I returned in time for Avi Bryant‘s talk. Quote of the presentation, paraphrased: “… and if you find Montréal cold, you haven’t been drinking enough. With brings me to career advice…” He taught us that bad programmers copy, good programmers steal. That is, take advantage of all of the research that we as computers science students have access to. Implement bleeding edge algorithms. And for God’s sake, don’t write it for developers, write it for users. He suggests that you read proggit and Lambda the Ultimate.

After that came Giles Bowkett‘s keynote. He’s a great speaker and kept the crowd laughing between bursts of profane insight. Unfortunately I couldn’t take notes, his 400 slides were pretty rapid fire, but it was recorded and will be posted. Until then you can watch his previous presentation at RubyFringe. The main thing I took away: Learn everything you can, even if it’s not directly useful it can expand your thinking. Don’t get bogged down in a large corporation if you don’t want to. Also, your small business is not your career, it’s part of your career.

Later on, we checked out La Banquise for poutine. I had a poutine kamikaze. The selection was amazing, they did not skimp on ingredients, and ingredients always consisted of far more than “gravy, fries, dairy product.” I had a forkful of cheese curd the size of a baby’s fist. The portions were gigantic, I couldn’t quite finish it. On the way home I was mystified by the selection of alcohol available in the local Mac’s.

I settled for the night splitting a twelve pack and trying to use the internet. Best Western only allows us 250mb per suite of four people, which as you can imagine does not go far in a room of software engineers. We’ve exceeded our quota every day, and last night we asked them to reset our quota. They did, sort of, by giving us a new wifi key with 100mb and telling us that it will not be renewed. We’re going to run out of wifi before the end of the today I expect. Very disappointing. The cost of transfer is so cheap, especially in Montréal compared to Toronto. I can’t believe Best Western is so stingy. They won’t even let us buy extra quota. They had downtime for most of the first day too. The average Motel 8 had better service. Maybe I ought to stay at one of them for my next conference.

The first presentation was a pep talk given by Leah Culver. She explained how software engineers can thrive under pressure with perseverance and creativity. She extolled the benefits of open source, explaining how useful it can be towards getting a job. She snuck in a jab at RoR at the last minute that got the audience grinning. Her answers in the questions period were at least as well received. Her formula for success is already publicly available and open sourced on github. She inspired a drinking game and we might have already seen its first playthrough. Towards the end of the question period, when asked how she handles all the attention she gets, she unwisely admitted in a room full of software engineers that she loves boys. Thank you Leah Culver, now we know how to be awesome! I hope that you manage to find some peace and can enjoy the rest of the conference!

I already posted my notes for the Radialpoint talk. The most important thing that I took away is that every software engineer ought to have a 1.0. That is, work on something new and unproven, and it’s best to do it early on when you can absorb risk much more easily. You won’t have a family to starve by working at a startup when you’re a fresh grad. I was surprised when I found out that Radialpoint used to be known as Zero Knowledge Systems after the cryptographic proof. I recalled Austin Hill’s blog and sure enough the two are related. Marty of Radialpoint was the first employee of the company that Austin Hill co-founded. Austin shared his endorsement.

The last talk that I attended was about the Lively kernel presented by Dan Ingalls. The website does a much better job of explaining what it is than I could this long after the presentation. After the talk I asked him my two questions. It turns out that there has been work towards a translation layer between Qt and the Lively kernel! Later on I’ll post pictures and dig through the mailing list for more details. Right now I’m too tired.

In addition to reading up on the Lively kernel, posting pictures and doing CSC 324 homework, tomorrow I intend to check out some of the recommended local cuisine. Good night.

The answer to that was yes. Dan explained that there was effort to build a layer for Qt and there was talk of building one for GTK. He wasn’t sure whether it was publcly available but suggested I sign up for the mailing list.

That answer made my next question partly moot. I asked why pick Lively over another web OS such as Eye OS. Dan explained he didn’t know about Eye OS in particular but offered the phrase “self sufficiency” and how the code could modify itself. Interesting. I didn’t see the examples he used in the demonstration, but it’s worth digging into.

Marty is the speaker. He’s from Radialpoint formerly known as Zero Knowledge Systems, was employee #1 and helped build up from startup to medium sized enterprise. Zero Knowledge based off the crypto proof.

Learning is essentially unlimited at the beginning of the career.

Innovator’s dilemma: the more successful a company is, the more weighed down it is by its own success. For example, Microsoft exists because IBM was too distracted with enterprise customers to worry about a personal OS.

Every engineer should have the opportunity to create a 1.0, create something original. Get working on that 1.0 as early into your career as possible.

Takes an awfully long time to go from an idea to a product.

Innovator requires determination.

Skunkworks departments: Big companies organize development into two departments. Maintaining a product that’s already into market. And creating a new product for products and concepts that are allowed to fail because they’re on the edge.

Or, go venture capitalist, it may be cheaper than having your own R&D department. The drawback is it’s really hard to integrate companies when you buy then, especially when you have little history of innovation and pick up a startup. Most talent will jump ship.

Or, manage a company differently at different times. This is like Radialpoint. See a saturation point for certain products, when it happens to core products, switch back to startup mode.

Cost of being on the edge: If it doesn’t bring in revenue at such and such a frequency, that arm will be cut off. This is what happens now during the recession. Doesn’t usually allow for internal transfers.

Microsoft is tied up and gives Google a crack at market leadership now. Take a look at Android versus Windows Mobile.

Strengths of recent graduate: Have an ability to absorb risk. Family won’t starve if you take a pay cut to create a startup. Much closer to the edge than you’ll ever be again. You _will_ be out of the loop eventually.

Questions!

If I wanted to be an intern at a security company, what are you looking for?
Not domain expertise. General knowledge about networking and security is fantastic. Read the Bruce Schnier books. Applied Crypto is very dense, but check out his other books. Principles of security are all in there. Design principles. Never give up on the technical questions. They never expect you to get an elegant solution, they want to see people work and create a hack, then refine it. Don’t worry so much about domain expertise at the intern level, it’s a general interest in software development. Ask yourself “Would I do it for fun or for free?”

Austin Hill co-founded Zero Knowledge Systems!

Watching websites
Presentation by bitcurrent

• Used to work for a company called coradiant
• Roughly one third of the audience monitors their website for uptime
• Startup 101: New idea -> Execution -> Feedback (did it work?) -> Money left? Yes: New idea. No: End.
• Do we understand our users? Is it easy and intuitive? Unfortunately usually this led to trial and error
• Internet makes us make mistakes faster, which isn’t necessarily bad
  • Most startups don’t succeed in doing what they set out to do
    • Amazon was just a bookstore
    • eBay sold Pez
    • F5 made hurricane modeling software

• Mistake speed is critical for the “money left?” stage

• Different sites need different metrics
  • Collaborative sites? Check comment rates, how engaged they are, how loyal they are
  • Ad supported? Click rate, buy rate, continued tweaking of campaigns
  • Which are you?

• Want to see what users do what we want
  • Enrolment, purchases, invitations, stickiness

• Want to see if the app is fast and reliable
  • Uptime, latency, SLAs, functionally correct, well maintained

• Understand our visitors
  • Intentions (we know what they want to do)
  • motivations (we know why they came to our site)

• Want to see if the app is easy to use
  • Easy to learn, clear and fast to use

• Our eyes: Synthetic tester, analytics receiver, passive capture, survey site
• Four big questions: What did they do? Could they do it? Why did they do it? How did they do it?
• What did they do
  • Domain of web analytics
  • Acquisition, usage and referral are important
  • Acquisition: what attracts them best?
  • Referral: what affiliates did they go to and why?
• Previously this was done by parsing access_logs, now people use AJAX to feed data back to the server. Example: Google Analytics.
• Set up goals and funnels for Google Analytics if you use GA
  • Much better to see what sites give you better conversion rates rather than bulk of traffic
• Don’t just look at numbers, look at comparisons to your site’s previous rates as well as to othe rsites
• Check out Clicky.com which gives better individual user look
• Also check out Mint
• Five levels of metrics: Check the slides, useful info but too much to copy
• These client side analytics have a problem with expert users. That is, users very familiar with your site. They tend to click through too fast, before the javascript at the end of the page has finished loading.
• What could go wrong? Check the slide for this too.
• Rather than “Oh? Something broke? We can give you a credit.” use “Excuse me, we’ve noticed that a proxy a x.y.z.a is interfering with the use of our site, here’s a credit and by the way your IT team can fix this by…”
• Use pingdom.com for testing
• Passive capture is the most accurate yet painful way to debug. Wireshark is free, TrueSight AIM is expensive but cool.
• TrueSight can tell you things like how many user sessions were interrupted by an outage
• Found that synthetic tests are often much worse than TrueSight, don’t show you the outliers very well
• Voice of customer: example of hotel booking site. Low conversion rate. Popped up survey, survey said that most users were just browsing to see the price and whether it was worth travelling at that point. Hotel booking site created new sidebar saying “Just browsing? Sign up for price drop alerts.” Conversions went way up.
• Also interesting: Clicktale. Lots of mouse and keylogging.
• Useful graphs: 80, 85, 90, 95^th percentile for load time over transactions per second
• Problem: AJAX! Standard off the shelf tools don’t see a prolonged pipeline containing 10 results as 10 results. Sees 1 very long request.
• Question period: Only question is security and how off-the-wire sniffers can read SSL. By giving them the private key.

Freshbooks talk
DabbleDB

- While doing dev for Viaweb, took heat from pseudotechs like VCs and industry analysts for not using a relational database.
- Used files
- also took heat for using cheap PCs running FreeBSD

Two companies that used nonconventional data stores: Yahoo and viaweb
- Yahoo: because we must. one large data set.
- Viaweb: because we can. masses of tiny data sets.

- Facebook does run on MySQL
- still using a cloud the way they cluster their database
- we’re not going to do the same thing

- Open for mere mortals as a cloud: AppEngine, SimpleDB, SSDS (from Microsoft) – presenter is not a fan of MS but still likes it

- Common for all of these: much more restricted feature set than typical relational database
- No joins, grouping, aggregation. Restricted sorting and restricted CPU. Dynamic schema.
- Done because at this kind of scale, doesn’t make sense to do these
- Anything that’s going to require these requires a query planner that can translate
- To implement these restricted ones .. some will be fast, search on index, some will be a linear search
- Linear scan doesn’t make sense for large database
- Restrictions are to make sure that your queries will execute queries

- Loads of concurrent requests are fine. Get results all at once.
- Have to be able to group queries yourself. Good library support is essential. May need to build your own.

- Must do a lot more work on write than you are used to doing at read. De-normalization (oh no)!
- Store lots of information in one data field?
- remember that there won’t be cascading on update

- On caching: Rarely need to use something like memcached since there are no expensive query. Problem is that writes won’t take for up to a minute. Cache insertions.

- Tips: concurrency is your friend, get good at grouping/sorting in local memory, compute on updates

- Data management that via web chose: Load machine with RAM. Have _all_ of customer’s data loaded into memory while customer is working on it.
- Issue with this is you can’t load balance the queries, because the live updated data will only be on one server
- Another worry: data loss when you kill -9

- Slowly write changes to disk
- Look up Prevayler, java library
- Inspired code in other languages
- Formalizes the pattern of keeping pattern in memory. Changes get represented in a Command object.
- Serialize the Command objects into a transaction log
- Checkpoint occasionally (write up whole state of the world)
- Replay the Commands when needed
- Things crash? Minimal data loss due to frequent checkpointing.

- Jotspot used Prevayler, he thinks

- Tips: fulltext index covers a multitude of sins, linear scans can often make up the rest, again get good at grouping/sorting in local memory
- Seems getting good at grouping/sorting in memory is essential

- Techmeme fits into RAM! 600mb of data. Why bother caching database when you can cram the whole thing into RAM?

- What about: Transactions, load balancing, data size > RAM size?
- Two answers
- Transactions: probably don’t need, try to get rid of concurrency. If you partition finely enough it might have five people per partition, one or two people on at a time. Mutex to execute the command object. Serialize rather than have optimistic concurrency.
- Load balancing: Not that big a deal. Don’t need a customer balanced accross many servers, need your customers partitioned over enough servers.
- What happens when someone wants a gig of data? Can’t put that into memory. (Or maybe you can? 16gb servers are reasonable.)

- Second set of answers
** – Demo of MagLev – new product hasn’t been demoed before
- new Ruby implementation built from the ground up for scalable web applications with persistence and caching built in