Soon (hopefully), the search results for places like Piece will have extra information like the number of reviews, the average rating and (where available) the average cost of a meal for 2.
Muxster is getting closer and closer to a stage where we want to start user testing and as part of the testing process we’ve set up a bunch of Mux’s for the various X-Factor finalists to demonstrate some of the features.
You should follow Muxster on Twitter to keep up to date with what’s happening and to let us know if you or your band are interested in doing some testing for us.
Update: None of these links work now. Now that Muxster has been launched we decided to remove the test accounts we set up and start afresh!
Again these are just the short(ish) notes I made during the conference on stuff that caught my imagination, made me think or make me want to do more research on.
Future of Frontend Engineering
Hasn’t said anything whatsoever.
Twitter Labs launching
Open source stack
Build an API NOW
Future of the Cloud
Transition
Confusion managements trust security transparency
Lack of transparency – need to know whos providing layers
Cost of not using vs risks of using
commoditisation is driving towards a service based economy
too many providers at the moment to allow secutriy and interoperability of the platform
ubuntu supporting ec2 api and have ubuntu distro that provides same api
standardisation will happen
cloud not green
enterprises not ready
standardisation will lead to more innovation
Rails 3 and the Future of Agile
Defer decisions to the last responsible moment
Rails makes it easier to experiment with other techs
Ruby Rails Agile – yay
Believe in your product. Best experience.
Not just founders and CEOs – whole company.
Consitency is state of mind
If you build it they will come is bullshit.
Be real. Be human. Admit mistakes – people will forgive you.
Qype have meetups – get their community involved – take them out and reach out to users – get all folk who’ve been to bars or vegetarian restaurants and invite them out.
New restaurant opens – email biggest reviewers. Then show this on the site?
Local as a platform. Needs to evolve.
Chilango had a buritto eating competition – awareness.
Create new identity from what your team values
The Future of Print Journalism
Fastcompany.com – launched on Drupal
Print – an elitist commodity – will cost more.
23 of 25 top news papers numbers decined
Craigslist and google killing print?
SodaHead – social news. Promoting news
Huffington Post – high FBConnect integraton
Digestable news
Glam Media – Tinker – apps for top conversations from twitter and facebook
Picure the impossible – newsy, get video for websites – service modes
Engagement new revenue streams
Where are kids viewing content?
Startup Metrics for Pirates
Keep it simple & actionable
fast, frequent iteration (& feedback)
measure conversion
focus on user experience
Progress is not equal to features
Acquisition
Activation
Retention
Referral
Revenue
Product / Market Fit – if we took a feature a way would it matter?
Focus on critical few actionable metrics
Focus on stuff you already have – be brave enough to kill shit features.
Kill a feature every week. Iterate on the stuff that visitors really love.
Make a good product
Market the product
Make money
“You probably cant save your ass and your face at the same time so choose carefully.”
Practical Advice for Managing the Growth of your Web App
Efficiency != Scalability
App will eventually become i/o bound – cpu, db, disk, netowkr
Scalability means overcoming i/o boundries in financial terms
Working on scability is always beneficial – optimisation is sometimes beneficial
Premature optimisation is the root of all evil
Steve Souder books
Smaller – focus on scalability
Larger – focus on optimisation
Process 250k feeds every 45min
Adding hardware to go faster is trivial (but costs)
Could be more efficient (but resources best used elsewhere)
Facebook saved money by changing to commodity storage systems – efficiency savings implemented once implemented
Hal Henderson – Building Scalable Websites
Scalable Internet Architectures
As the title says these are my raw notes that I jotted down in Evernote while listening to the presentations at Future of Web Apps London 2009 on day 1.
I’ll need to distill them into something meaningful on Monday for my colleagues to read but for now they’re here in all there messy glory.
Anything thats missing I either 1) Didn’t hear anything interesting in or 2) Had arrived late due to travelling from Glasgow in the early hours so missed .
3 Vital Marketing Items
Don’t build what you don’t need to launch.
Tracking system – build one right at the beginning to ensure you know exactly whats working for you.
Sometimes GA cant give you as much data as you need so build custom metics to know exactly what your users are doing.
Future of Javascript Design Patterns
“meta programming is like trying to do crack cocaine responsibly.”
Use relevant frameworks – simpler tools for specific reuirements.
small taks
seperate business logic
Checkout:
Dean edwards Base
Dan Webbs Low Pro
Avoid over abstractions
Trust your code – dont bloat with error and type checking.
Passion and Paychecks: Opensource
Communication
sell brain trust
community
free time built in
clients pay top dollar for passionate people who know their shit
work atmosphere makes people not want to leave – lullaby. find good passionate people and pursue them
passionate people – hiring people not positions
keep people passionate or they’ll leave and go somewhere else
let people kick ass
Atlus
280atlus.com – $20
Cappucino – server agnostic
Cappucino server – running javascript.
RESTful interface for accessing server side.
Go Niche, Get Rich and go Mainstream
Revision 3
New generation of consumers have abandoned tv and traditional media.
Digital natives.
IPTV – what you want, when you want it.
City Sourced – allow users to report problems about a city. GPS location, direction, photo – crowdsourced IPhone app
Low budget & targetted approach – lower cost/ higher return – hit evngelists
Boosters size -> rail -> trams -> wagons -> chariots -> horses asses
Innovation enables new monetisation models
Social networks will redefined ecommerce
Mobile ecommerce – unprecedented takeup of ebay mobile app – $350m paypal transactions through it in the last month and a half.
Paypal platform launching November 3rd. Send money, payments etc. paypal.com/innovate2009
Web App Marketing Strategies
Spymaster?
Build a great product
Passionate Community
Engage with the community
Incentivize users to market for you
Empower your heaviest users.
Usual stuff
FB Connect
Registration
Interaction
Social context – friends
Fan box, comments box, live stream box
Live feed on big news items? Live stream on live tv shows?
traffic – share link
engagement
registration peronsalisation – activity of friends – what have your friends done?
Huffington post traffic referrals increased 300%
Translations app released
Wizard for setting up connections.
Typepad
Motion – like an easy to drop in twitter (js based) Open source – for building communities. Developer.typepad.com
Run on Django / Python
University session tomorrow.
HTML 5
Yawn
Guardian – Stacks for a mutualised newspaper
Mass media does not have bi-directionality
Be part of disruption
Mutualisation – linking consumer with the journalists – everyone has a means of production
Consumers asked to provide footage thats used to generate stories – reaching out to audience to help co creation.
Co-fabriction and co-distribution – XML JSON ATOM of over 1m articles. Datastore – excellent datasets that can be used- weave the guardian into the fabric of the internet data.hmg.gov.uk – 1000 datasets
People make amazing things with open data
Promotion of the Guardian outside the uk.
Guardian app gallery / store – community
Guardian data blog
Flickr group showing data use
Django framework – scaleable – used for Guardian app for analysing expenses
Rapid turnarounds
1 dev / 1 week
1 designer / 2 days
Deploy on EC2 for limited time
Collating conversations of journalists
AppEngine task queue
Guardian as a platform not just a publisher @jaggeree http://www.slideshare.net/openplatform/building-the-stacks-for-a-mutualised-newspaper
I mentioned previously that I’d been looking at Amazon EC2 as a solution to host a site that underwent heavy bursts of traffic for only a couple of hours a week. The expected traffic bursts were 10-15k visitors in a 2 hour period and unfortunately a small EC2 instance just wasn’t enough. The next trial was with a large EC2 instance, with Apache tweaked to allow more simultaneous connections. Results went better than first time but I ended up having to bounce Apache a couple of times during the time frame to get it back up and running.
So fast forward to this week and it was time to try again. This time I went with an Extra Large instance running a stock 64-bit Ubuntu 8.04 and Lighttpd instead of Apache. And this time there were no problems. For the full 2 hours the site remained up and responsive, and because I could time the instance to come up just before the additional cost was negligible.
To give you an idea of the kind of traffic it had to deal with, the site took in just over 25,000 unique visitors in a 2 hour time frame. Glad to get the problem solved and now I know what to do next time I need some heavy traffic handled.
The actors from the STVJobs.com commercial performed their ‘Haka’ jobs dance today in Queen Street Station at noon. They also did it in Edinburgh in the morning and Aberdeen in the afternoon.
As mentioned previously once of the last few things I did in my previous position at ocean70 was to migrate their existing dedicated server over from a 2 year old Fasthosts dedicated box to an Amazon EC2 instance. That was a little over a month ago now so I thought it would be interesting to take a look at the cost for the set up in comparison to the dedicated box used previously.
To put things in perspective the box we were replacing was a 2 and a half year old Fasthosts DS 400 running Fedora 6 clocking in at £69 + vat a month (£81 in total). In the box was a 2.8ghz Pentium 4, 1gb of Ram and a 120gb IDE hard drive. At the time this was enough for our needs – but despite regular backups and an in-house server available for failover there was always the chance that this would go down and we would be in a right mess. Luckily that hadn’t happened and time was right to move our hosting to EC2.
One month has passed on EC2 and i wanted to take a look at how much the server has cost to run for a month. Total charges clocked in at $125.46 (including taxes) which at todays exchange rate equates to £84.77. Not too shabby for a vastly improved infrastructure.
To actually have the AMI running 24/7 for a month (and remember, January is a long month) clocked in at $81.84 (£55.30) while the transfer in of 23gb cost $2.30 (£1.55) and our outgoing transfer of 25gb cost $4.27 (£2.88). To put the bandwidth costs into perspective the instance is home to about 8 or so websites receiving fairly low traffic (max 100 uniques a day each). The only other significant cost being the 120gb EBS block which cost $15.00 (£10.13) for the month.
So thats a £5 a month increase for a system whose hardware will never date. Can be brought back up in 5 minutes should anything go wrong. Has virtually limitless storage facility, and on top of that is a hell of a lot faster than its predecessor. Bargain. Will be interested to see how it averages out over the year, especially as a few more sites are put on it.
Two of the projects i worked on at oceanseventy, a project management application and a photo management / database have been made open source. You can find epicentr and efstop over at Google code.
Epicentr is in better shape at the moment – both applications had to be stripped of some proprietary code that couldn’t be open sourced – but i wouldn’t recommend installing it until the initial release.
The great thing about these projects going open source is that myself and the other contributors can finally move them in the direction we want to and not be hampered by the ‘business’ needs. Plus I can continue to work on them now that i’m moved job.
As I mentioned previously, we (oceanseventy) decided that recently to take advantage of the slowdown around Christmas and start to migrate our hosting set up away from a Fasthosts dedicated server.
For 2 years we’d been using a Fedora 6 box which had been running fine, but it was becoming difficult to keep the OS up to date given Fedora’s habit of ignoring previous releases fairly quickly. Other issues, like the fact that there was a single hd in the machine which could potentially die at any time and cause much pain, prompted the desire to move everything to something more robust, flexible and easy to backup. Enter Amazon EC2.
Fedora was ditched in favour of Ubuntu 8.04 – the latest LTS release – and within an hour we had set up a small instance in the Euro availability zone. The doc’s for EC2 are excellent and take you through the process. An Elastic IP was attached to the instance to ensure we could maintain the ip should the server ever go down.
We also attached a 120gb Elastic Block Store instance, mounted at /var/www to make it easy to take all data easily to another instance should the need arise. Backups are handled with the AMI and EC2 tools. Nightly the instance is bundled and stored on S3 and nightly a snapshot is take of the EBS block as well.
So far the instance has been rock solid reliable and much faster than the old Fasthosts dedicated box we had. EC2 is exactly what anyone hosting websites need at the moment – no tie-in to any physical hardware and the flexibility to scale your hardware as required.
I’ve seen a lot of criticism about hosting web sites on EC2 but almost all of the issues I’ve seen are no longer relevant as Amazon have taken steps to improve EC2 and add critical features (like persistent external IP addresses).
Really glad I got the opportunity to implement this as my last act at oceanseventy. I’m definitely and EC2 advocate now and hopeful the skills i’ve learned will help my in my new position at STV. Hopefully i’ll be able to use EC2 for my startup project as well.