IanOzsvald

I spent today (Thursday) talking with the Executive Director (Horacio) and a few of the senior staff members. I presented this morning on Lean Startup processes for the newly-arrived.

I'm in Round 2, I've been giving feedback for months to the staff. The feedback I got today about changes they're making during Round 4 (for those just arriving now) is pretty impressive. If they carry it all off then the programme will be improved by a fair margin for subsequent Rounds.

The improvements are still under discussion but ought to address many of the mentorship/fund raising/expectation issues that have been raised (at least - I have cautious optimism that that's the case).

Disclaimer - I don't work for SUP, I'm almost finished here in Round 2, I've had my complaints during the programme but generally they're reacting/improving to the significant issues. Things are improving for every new Round.

Me again (Round 2) - I spent 2 days doing the first reimbursement (following instructions, taking advice from Round 1 people) and then 1 day or less for each subsequent monthly meeting (including having the meeting). Write-up from a few months back: http://ianozsvald.com/2012/03/24/this-week-in-startups-stron...

All in I'd expect 1-2 days per reimbursement meeting (including meeting time), less as you get better at it, less if you reimburse just a few big things (rather than lots of small stuff). Cash payments require less proof (less paperwork) than debit or credit cards (CCs require most). That's about all there is to it.

The Round 4 application process was discussed here a week back: http://news.ycombinator.com/item?id=4158967

It has discussion from Round 2+3 people on the pros and cons of the programme. Sidenote - I knew Liis (she's lovely), her concerns are echoed by others (including me). My take on the programme is nonetheless positive, you just have to come with eyes-wide-open.

To clarify the theft issue (I'm Round 2, almost finished here now). At the start the co-working CMI office was safe, we guess that no 'regular folk' knew about it.

Now it is better known and there's no access control (there are security guys on the doors, but they're not really checking us). Several times in the last few months different strangers have walked in, spotted a lone laptop and walked off with it. The same thing has happened in restaurants.

This is petty theft, a pity in a co-work environment but given that it is large and open this is somewhat unavoidable. I wouldn't leave my laptop unattended (without asking a neighbour etc) in a London co-work space either.

In general I feel very safe here (really, no problems walking home for an hour at night, using the metro etc).

Certainly coming here I've realised that I ought to consider non-English languages (particularly Spanish) for web apps. Most folks here in Chile don't speak much English yet their middle class has a growing disposable income. LatAm's power does seem to be growing. Also some of the companies in our rounds (Round 2+3) were quite inspirational - water purification, classroom neuroscience, human-waste->fertiliser solar cookers, solar powered food cookers, wind farms, safe remittance systems for migrant workers. I'm most curious to see what Round 4 brings (quad-copters from one I think). Accelerator high-tech is more than Web2.0.

I should clarify - if coming from the UK you'll want more cash than if coming from the US. In the UK the Chilean work visa costs £1220 (the highest in the world from memory - set by the govt., nothing to do with startupchile). Flights were also over £1000. Coming from the USA will be much cheaper.

@jot (from Round 1) made some great comments a few months back: http://news.ycombinator.com/item?id=3810821

I added a few notes too (I'm Round 2) - definitely think more along the lines of £10kGBP (about $15k) if you want to have a smooth ride. Two months from landing to getting money in CLP into your Chilean bank account (it'll only go in your new Chilean account) is a minimum. Sending money aborad is a pain and requires bank visits (online banking here isn't brilliant).

There's a comment on this post about Mentorship Groups - I setup the first (as best I know) which we've been running for 3 or so months, details here if you want your own: http://ianozsvald.com/2012/05/16/mentorship-groups-in-startu...

The lack of mentorship will apparently we changed for Round 4 (fingers crossed). Definitely come here knowing what you want to achieve and take all collaboration/support as an unexpected bonus.

Overall I rate my experience here as positive, it certainly got me out of a rut back in the UK. The participants and staff are supportive (and generally lovely), the push to make us achieve is rather weak (hence starting our self-mentorship group). The programme will improve with each new round.

Just to note that this was done by our intern for our AI platform, its a nice little library that automates a badly documented training process (Python).

In addition, here are 3 blog posts which might help:

http://ianozsvald.com/2012/02/20/startupchile-pycon-strongst...

http://ianozsvald.com/2012/03/24/this-week-in-startups-stron...

http://ianozsvald.com/2012/03/30/2nd-dataai-meetup-official-...

The third is about the AI/data meet we setup - over 100 folk turned up to hear 4 of us speak at the just-opened hackerspace, a founder of Skype also spoke and a local entrepreneur brought local beers. It was a darned fine evening.

Our padrino has been ace - we drink with him every week and he drove me to the FIDAE air show last Saturday. We seem to be an exception, many others only meet their padrino(/madrina) once or twice. As ever if you put time into a new relationship with a stranger there's a chance it'll pay off, you have to work at it (and accept it might not work out).

As @jot's friend, member of Round 2.5 and Emily's other half I'll add: reimbursements took 2 days for us the first time and 1 day the second time. The first time it took about 20 days for the money to come back, the second time about 10 days.

The stories from Round 1 (e.g. @jot) scared the bejeezus out of us, we spent a solid Sunday getting every scrap of paper to prove our expenses (from invoice to credit card to bank account to our names on a bank statement proving ownership). They're more on top of the process now than in Round 1, we did have to get some extra docs but it just involved a few phone calls back home.

I believe for Round 3 it is a bit easier and I'd expect it to keep improving. Assume 1-2 days admin on this every month regardless (just with less follow-up stress after you submit your docs).

@jot's a friend of mine, we (http://strongsteam.com/) followed out in January in Round 2.5. His notes match my experiences and those of others here in Round 2.5.

A group of us have just started a weekly self-mentorship group, frankly I should have started such a group 2 months back (I've been here 10 weeks now). Get on with this quickly when you arrive (and expect to iterate, experiment and change all the time)

Definitely come with spare cash - $10k is probably about right. Better yet - have a client lined up before you come who pays you as you hit your deadlines whilst here. It kept us on our toes, we delivered, we had cash-flow.

Our round (Round 2) and @jot's (Round 1) each had 100+ companies, I believe another 100 arrive this/next week with Round 3 and we'll expect a similar nbr for Round 4. In Round 2 & 3 there were 600 applicants with properly-completed forms, if you're interested in the odds :-)

Personally I'll say that being here is very nice, it is great to avoid having to take on client consulting work to pay the mortgage. Things could be better (reimbursements, mentoring, access to more money) but things are a lot better for me than if I'd stayed back in the UK. I'm going to meet 1,000 pro-active folk out here in the following 5 months, I'd not find that many in the UK via my old networks.

I agree that many people here haven't launched in a public way. We ( http://strongsteam.com/ ) have some closed alpha users and add more every few weeks, we'll be months before hitting beta. There are however a nice set (10%?) who have launched and may not be in @jot's page, some are physical products (e.g. photo booths for bars, waste water boilers->clean water, human waste sanitisers) which don't appear on the list. There's quite a mixed set out here which makes for very across-the-board conversation.

I'll echo @jot, we're another UK team (Round 2, been here 8 weeks). The pricing is similar to Brighton back home (smaller than London on the south coast), we get a 2 bed very nice apartment for £850UKP equivalent (though we pay a £100 excess each for that, no pool here sadly either), our bills for cooker-gas/electric/internet are another £100/mo on top. We could have chosen more central, smaller and cheaper but we chose a nice/larger place. The Metro is roughly £1 per journey (bus/metro/both). Cell SIM is £20/month for HSDPA with a reasonable data limit. Food is similar to back home (lunch costs £10, evening meal £10-£30, supermarkets the same as home).

Basically we get to live in a city for the cost of a small UK city which, for us, is quite a nice deal. Nobody else has complained about the amount of money (which comes for 0% equity), the problems are more around the govt. bureaucracy and the refund process which saps time. We jut got almost 100% of our requests first month reimbursements back at a time-cost of 1.5 days.

At the end of the day we've got $20k USD to invest in people and $20k to live off here. Fine for MVP->launch (we're 3 weeks from alpha launch with http://strongsteam.com/ ).

'excludes' scipy in that the numpy C API won't be available so there's a massive hill to climb (as best I understand it) to support most of scipy. Re. getting the pypy team to do the whole port - agree that'd be entirely unrealistic! Re. initial port - since there is more than one way forwards and I want to financially support the project (having pledged £600 earlier), I'd like to understand the risks and benefits in the options.

I spoke with Travis on Friday at Enthought and he opened my eyes to some of the possible problems of the pypy-numpy effort (which I've started discussing: http://mail.python.org/pipermail/pypy-dev/2011-October/00860... ). I hadn't realised that the port might exclude use of the rest of SciPy, this strikes me as a massive missed opportunity if it comes to pass. Perhaps a few other knowledgeable folk could post in the pypy-dev thread so we can reasonably understand the limitations and possibilities of the various approaches?

One obvious point - startupbritain is about general startups (there's no requirement for tech-only startups on their site). @jot's jfdibritain has a decidedly tech focus, it doesn't help with e.g. manufacturing (when someone wants to make something physical), retail (regular selling/buying like most of the country are used to) or other conventional routes. Maybe the sites aren't trying to achieve the same thing?

As rwmj mentions below - avoiding the sums that sanity check the business could be an expensive mistake. Doing the sums is easy (and if you can't make up rough numbers, you ought to ask if you really know your domain). I'm not advocating a full on business plan at the start - just a day spent roughing out the key numbers. Are there enough customers? What do they pay competitors? How much must you sell to break even? Do you know where your customers hang out? That said - if your business has a significant cost of failure (e.g. buying big plant equipment, renting a lease, manufacturing, big design costs etc) then not doing planning is probably rather foolhardy.

Follow the Inspiring Stories under Getting Inspiration - most are not tech-startup focused. Also look at 'About Us' and read the bios. I'd argue that 'startup' for the government relates to any new small business, not just a tech/web focused one.

Here's a further demo showing a social graph (Twitter friend relations at a BarCamp) with an image hack on ProtoVis' directed force graph: http://ianozsvald.com/2010/10/05/visualising-lanyrds-social-... It works great in Chrome/Safari and ok in Firefox. It runs like a dog on the iphone 3G. Kyran (who hacked ProtoVis) is working on a Canvas reimplementation which runs faster on mobiles.

Headroid's author here - yes, the background skills took a fair while to acquire. As a kid (I'm 34 now) my late father (hardcore electrical engineer) used to bring home electronics junk so I could build robots aged 10 (solder+Commodore 64 joystick port+Lego+Basic). Then I got bored of robotics for a long time. For 11 yrs I've worked as an A.I. consultant after an AI degree/MSc, recently I wanted to hack, play and build fun stuff so I setup the AICookbook.com with Google Group and I'm posting my play-time projects. Headroid is one of them, there's also OCR (see the Plaque entry on the blog), NLP, recommendation engines, data mining, game play (auto Zork solver anyone?) and more to be done. I'm also a part of Brighton's HackerSpace (BuildBrighton.com) which is where I met BotBuilder who built the servo board and frame. All are welcome to come join the fun :-)

To see ie6update in action watch this 3 minute screencast: http://www.youtube.com/watch?v=TclHo_F7rsE

I wanted a demo but didn't have IE6 to hand, in the end I installed ies4linux -> Wine -> Jaunty RC and I figured that others might want to see it in action so I sat up until 2.30am making the screencast.

Nick says he'll integrate it into their blog shortly, but in the meantime...

+1 for CamTasia (Windows) and ScreenFlow (Mac), also RecordMyDesktop (Linux). CamStudio (free Win) ok. Online solutions (screencastomatic, screentoaster, jing) also good. I'm a professional screencaster (http://procasts.co.uk) - if you want to crib examples, see my Examples page, they all reflect live use of screencasts that make users happier, convert more visitors, reduce support costs etc. Definitely get a good mic - I prefer pro audio equipment but you'll go a long way with a decent USB mic (not 3.5mm jacks, they get electrical noise from the motherboard). If you can't do good audio yourself then see something like voices.com for outsourcing but melding the video+audio can be problematic. You do need good audio, as others have said the easiest way to kill your presentation is muffled, noisy, volume-changing hard-to-understand audio. Expect 1 day to do a first reasonable video of a few minutes, several days if you want a perfect cut with great audio, a bit more if you want to add effects (captions, fades, highlights, scrolls). Look at ShowMeDo for 800+ examples of screencasts made by many users on different OSs with different tools. Jolt Magazine has three useful articles: http://joltmagazine.com/2008/10/08/lets-make-a-screencast-th... http://joltmagazine.com/2008/09/23/lets-make-a-screencast-pa... http://joltmagazine.com/2008/09/17/screencasting-an-expert-r... See LongTail video player (http://www.longtailvideo.com/players/jw-flv-player/) to embed flv/mp4 result, they have great plugins, it works with thickbox/lightbox, stats tracking possible, lots of features, I use it for all my work.