Which made me (and a whole bunch of other people) joke about writing a filesystem to use it as cloud storage.

Then, of course someone _did_ it ,and someone else too. (Personally I think “encoding” your data into text chunks in the png file is cheating – surely storing it as image data, either data-as-images-looking-like-noise or data-steganographically-embedded is more “in keeping with the spirit” of using Flickr’s storage?).

Then @emailpdeluca asked ”so is raid supported with multiple accounts”?

But I was busy at work. so, of course, by brain decided to think about this and nothing else because productivity! - so…

A new plan.

Presenting

Washoe LAPSSFS

Washoe LAPSSFS is a distributed, fault tolerant, secure, highly available, photo storage provider independent – cloud storage system. It stores data in an “album grid” consisting of a collection of albums stored on free or inexpensive photo hosting sites. Each file you wish to store is first encrypted, then broken up into chunks and each chunk is stored in a photo on any of your configured photo storage sites/accounts. The splitting into chunks is done using erasure-coding, which means you can reconstruct the data even if one or many of the storage services become unavailable. It can also ensure that any one storage provider doesn’t hold enough encoded chunks to decrypt the file – even if they acquire the decryption key.

How it (ought to) work:

• Embed (or fork) Tahoe LAFS
• Modify/configure Tahoe to create chunks small enough to me encoded/hidden in photos (ballpark guesses – if we steganographically encode our encrypted data into the 2 least significant bits of a 16bit 2 megapixel image, we’ll be able to store a block of 256KB in each photo. This is a profligate waste of storage resources, but hey – Melissa just paid $1.1Billion for Tumblr, looks to me like they can afford it…)
• Embed those chunks of data into individual image files (either from a user-supplied image library, or by searching online image sites for CC licensed photos).
• Modify Tahoe to store urls/credential of hosting sites and photos as the location of each chunk – and to use those urls/credentials to retrieve specific chunks as required.
• Upload resulting image files to the available set of hosting sites/accounts.
• Profit

Future plans (for someone else to implement:

• Add the ability to proxy photo storage/retrieval requests for friends/colleagues/anonymous-random-people. This adds “noise” to any storage account traceable back to you – as well as hiding critical chunks of your encrypted data in storage accounts not related to you.

Development roadmap:

• I plan to post this rant to Twitter, and fully expect to see someone on Hackernews has it up and running a forkable on GitHub by the end of the weekend.

Wow! Welcome HackerNews visitors! Discussion here

This is a story about some ideas, some pivots, and one friend’s 3 month or so run at executing one idea.

From this

to this

TL;DR – Ideas come quick. You can have dozens of them a day if you’re in a creative mood. Execution is, for many ideas, orders of magnitude more effort. Any big, hairy, ambitious idea you have is probably going to require skills and experience you dont have, and you’ll need to find people who do and convince them to help or pay them to consult. That’s why incubators (somethimes) work so well. That’s why “skunkworks” projects in big companies (sometimes) work so well. Everybody “knows” this – here’s a story illustrating it…

This might be partly my fault.

Early this year, I got very frustrated on RasberryPi’s launch day, wasting 3 or 4 hours pounding reload buttons and having form submissions fail trying to order one. I eventually got an order in to Element14 (which is what Farnell call themselves these days in Australia), but it was already “on backorder” by the time I’d got their website to accept my money. So my instant gratification desire wasn’t fulfilled, and when it finally shipped later in the year I’d got busy with work and with other projects, so it sat in my pile of “stuff to play with later”.

At some stage, my friend Mark Pesce (and if that sounds familiar, perhaps you’re old enough to remember VRML and the Labyrinth demo from the late ’90s) mentioned he was waiting to get his hands on a ‘Pi, with vague ideas about teaching kids to program, kid’s robotics courses, TV shows – I’d taken him to our local hackerspace Robots and Dinosaurs to meet some like-minded people and see what gear and skills and people were doing hardware/software hacking locally. So I lent him my RaspberryPi to play with.

Shortly afterwards, he was tweeting about having an LED hooked up to the GPIO port, and a Python script running from the webserver on the ‘Pi, which let him turn his LED on and off with a cheesy little JSON API from anywhere on the internet. So, of course, I made it blink on and off hoping to annoy him, as you do. Then I made it blink … —- … (which is SOS in morse code). Not satisfied with that, I glued together a bunch of Perl modules from CPAN, and slurped up all Marks tweets and responses to @mpesce from Twitter in real time, and made his LED blink them all out in Morse code. (Partly ’cause it was fun, but probably at least a little bit because I should have been filling out timesheets or invoicing or writing reports at work, you know those most creative of times…)

A month or so later, he’s gone from a bare Raspberry Pi sitting on his desk with a single orange LED, to a (as he puts it) “pipebomb looking” contraption with about 64 individually addressable RGB LEDs zip tied onto a white foam base with a ‘Pi hanging off the end of it, which lets him play patterns of colours and brightnesses, all under  control from his phone via the internet.

This is where the real work started…

In early October, I got invited to a launch at FishBurners (a Sydney co-working space / incubator). Mark had put together a team of experienced and talented professionals, Kean who I knew from Robots and Dinosaurs and who runs an embedded electronics / microcontroller consulting firm, Robert from Tiller Design who I knew by reputation as the guy who did the industrial design on the beautiful Otto Espresso Machine, Breton coming up with awesome demos and writing 3 dimensional graphics abstraction libraries, Kate doing marketing and social media. All of those people are “worth” well over $150/hr, in the sense that they’d get paid at least that much by other businesses for their time. There’s certainly more people who should be on this list too, I know I put Mark in touch with Colin for information about getting a security audit done on the internet-facing hardware/software, somebody talented did a bunch of web design and WordPress configuration for them, the open and transparent financials on their website explaining the costings and the resulting Kickstarter target amount must have had some procurement/logistics/manufacturing experience behind them, there’s been clearly professional photography and video shot many times.

At the FishBurners event, there were close-to-production-ready prototypes of the light.

All of that had gone way beyond the initial “idea”. There’s been work put in by a dozen or more people, work witch might well have cost several hundred thousand dollars at full commercial rates. I’m pretty sure most of those people involved have only charged nominal rates, if anything – I’m sure quite a few of them have dug deeply into their own pockets to pay for the expenses of getting this far. Instead they’re all caught up in the excitement of “the idea” and doing hard, valuable, experienced, professional work; with the expectation of future payoff when “the idea” becomes “the successful business”.

This is “the execution”. The bit that’s a lot less fun, and a lot more personally and financially costly that “coming up with ideas”. And Mark’s team are killing it!

(and now, the “ask”…)

And they need your help. To do this right will take a lot of money. Go have a look at their Kickstarter page. If you like the idea as much as I do – chip in to help. If you can’t afford to, maybe forward it on to your friends – ask for a light-pledge for christmas. Tweet about it. Blog about it.

Go on. Help make a smart light exist…

A recent TechCrunch article by Jack Krawcyzk titled “Your Mobile Phone Is The Least Social Device You Own” had this graph near the opening:

Session length by Media/Device type (from a techcrunch.com article by Jack Krawcyzk)

It’s suggesting that people visiting your site on mobile only do so for a very short time compared to a “desktop” visit.

There’s a certain “well, _obviously_” simplicity in that revelation, but there’s also some consequences if it’s true[1].

On the Mightymedia website, we know from tracking webserver logs and Google Analytics that visitors are 85+% likely to be looking for one of two things – our phone number or our address. We’ve got an “About Us” and a “Services” section in our mobile site (you can see it by going directly to http://www.mightymedia.com.au/mobile even on your desktop browser), but almost nobody ever clicks those links. People hit the homepage and either click the phone number autodialling link, or go to the Contact page to see our address. (Or they bounce from the home page…) We had a mobile optimised portfolio page there for a while, and in 4 months not one person clicked it (except for us testing it and showing it off).

Your users goals when browsing your website on their phone are very likely to be extremely straightforward. They have their mobile phone in their hand, and they’re looking for information about you. It’s almost certain they want to call you or visit you. Make sure you’ve got you phone number, address, and opening hours really easy to find. That’s probably what nearly everybody is looking for on your mobile site.

Google breaks down mobile users into 3 categories: A. “Repetitive now” B. “Bored now” C. “Urgent now” (see: here). Mightymedia’s mobile website visitors are clearly all in category C “Urgent now”. You business might be different – Yellow Express (or their direct mobile site link) have “Repetitive now” clients who’re tracking parcels, SMB Interior Design ( mobile link) has quite a lot of visitors spending many minutes browsing their gallery, presumably in the “Bored now” category. But both of those still have very important contact details very prominently displayed.

Make sure you know what visitors to your mobile site are looking for, and what business goals you need them to complete. (And, as always, Mighty Media would love to help you develop your mobile website strategy, call us on 1300 050 690 to chat)

[1] The truthfulness there is not easy to measure. First thing I did was go and look in Google Analytics to see Average visit duration for mobile vs non-mobile visits across a bunch of sites. Unfortunately that doesn’t immediately show a much set of stats “proving” the assertion, with different sites and different time periods on the same site giving contradictory and inconclusive numbers. I think the problem is finding

If government departments are setting up mobile websites or mobile optimised websites, that’s surely just going to increase the number of people expecting a functional mobile website experience from every website.

(And, of course, if you’re looking for mobile web development in Sydney, get in touch with me at work at Mighty Media.)

torvalds commented 6 hours ago

On Fri, May 11, 2012 at 1:48 PM, Dominik Dabrowski wrote:

You might have fun raging on the internet, but I think your goals would be better served if you expressed your thoughts in a clear (maybe even polite) manner that doesn’t embarrass the people whose actions you’re trying to influence.

Umm. I think I’ve been able to reach my goals on the internet better
than most people.

The fact that I’m very clear about my opinions is probably part of it.

If people get offended by accurate portrayals of the current state of
github pull requests, that’s *their* problem.

I hate that whole “victim philosophy”.  The truth shouldn’t be sugarcoated.

                        Linus

(Go, read it. I’ll wait…)

Personally I think it’s confused on the important differences between tablets and smartphones, and also the difference between mobile websites and mobile apps.

In spite of that it’s a good read with some startling figures, including:

The Commonwealth Bank says 30 per cent of its online transactions now come from mobile devices, compared with just two per cent two years ago.

and

More than 25 per cent of tablet sales in the year ahead are expected to be to companies for their employees.

That second quote interests me mostly for what it’s not saying. Up to 75% of tablet sales are not being purchased by companies.

What are all those personally bought iPad / Android tablet / Windows tablet owners doing with their high tech toys? Have you checked your website analytics to find out? (Here’s a hint, a lot of them are spending incredible amounts of money…)

A fine sentiment, but shouldn’t _everyday_ be “in support of web standards”?

(Well, except for days I’m out riding the bike. Every _other_ day I care deeply about web standards – when I’m halfway through the twisty mountain roads, whether you’re using Flash or IE-only javascript or whatnot is the furthest thing from my mind…)

There’s a great post here about measuring visitors video performance – do you know how often visitors end up with pauses while rebuffering? Should you be compressing your video more? Could you get away with higher quality video? How can you even find out?

Go read Simons post – it’s a bit technically specific to his Omniture SiteCatalyst analytics setup, but it asks some good questions and provides ideas about how and what you can measure to monitor your site’s video playback performance.

I use two of those free services as part of our website strategy at the web design firm I work for in Sydney. mon.itor.us is great for tracking uptime and response times over the previous two months, and www.montastic.com have a really good email notification service. I recommend both of them.

Anyone who has a website needs to decide if they care about downtime, and if so, how much. Unless you don’t care at all, you need a way to measure it. And you probably ought to care because there are lots of implications of unscheduled website downtime, from the appearance of poor customer service to the SEO (search engine optimisation) implications of a flaky or slow website. It might not matter if your pictures-of-cute-kittys website disappears for a few hours or days (but I’ll bet the icanhascheezburger.com people know exactly when their site is down), but if your website is a reflection of your business you need to ensure it’s reflecting the image you try very hard to portray in the rest of your professional life. You wouldn’t leave you shop/office/showroom/studio closed up and not answer the phone at times your clients expect you to be available, right?

Ask your web people if they’re monitoring your site for downtime. If they’re not, ask them if they can. Or use one of these services to do it yourself.

Call us at Mighty Media for advice, or for a free “Website Health Check”, or let us help you with full a website strategy consultation.

Remember, if you aren’t measuring it, you can’t improve it. How is your website performing now? How can you make it perform better?

SEO In A Nutshell

People treat SEO like it is black magic, but at the core it is very simple: Content + Links = You Win.

A good explanation of “content” and “links”, and how to generate both. Some interesting numbers on the costs and profits from targeting long tail search keywords. The post is aimed at people producing and selling software, but the ideas and principles are applicable to any website…