prohacker.co.uk

professionl hacker, maker, developer and tech enthusiast

[prohackeruk-blog]

twilio

(quick post today, i’m in a hurry)

twilio is an online service which allows you to create programmatic solutions for various types of communication -- most notably, by text messages and voice calls.

with all of the alternative, less expensive methods of communication that exist today, i wouldn’t have thought this would be such a lucrative service, yet it’s something i’ve had several opportunities to play with recently since there seems to be such a large call for it.

what i leave myself wondering is whether this is just a case of the good old-fashioned telephone coming back into fashion for some unknown reason, or if, yet again, the attitudes of people in the business world haven’t been moving as fast as technological progression

[prohackeruk-blog]

trendlistr.com

i have a big love for small enterprises. a lot of respect should be afforded to people who are willing to take risks in order to run their own startup companies, and the culture that surrounds startups is pretty cool in and of itself. it’s a bit strange, but i’ve always liked it a bit strange. lots of interesting things happen when people decide that there’s a problem that needs to be fixed, and they’re going to be the one to fix it.

that being said, i’ve never been terribly interested in the standard ‘startup’ model myself -- probably because the politics of the situation doesn’t really interest me. a pattern we’ve been seeing for years now is the tendencies for anybody with a good idea in the tech industry to be bought up by the giants. it’s not necessarily a bad thing, either; it means that people who otherwise wouldn’t be able to manufacture or provide service on a large scale now have access to the resources they need, but let’s just say a six-figure salary and a company mansion aren’t exactly on my bucket list. ;)

this is why it was so appealing to me to work on a project that had very little to do with technology. trendlistr.com is an “online fashion marketplace for pre-loved clothing”, as described by louisa rogers, the m.d. and owner. that’s how i’d describe it, too, and i think it’s an excellent concept, so i wanted to work on it as soon as i heard about it. i don’t really know much about fashion aside from a certain someone, so i thought it’d be fun to throw myself completely out of my confort zone. the next month, when we were both at rise and design again, i sat down at a table with louisa and had a very interesting discussion about her project. she agreed to hire me, we worked together for a few weeks (this is actually the less interesting part of the process, at least for a blog post), and her site went live into beta about a week ago and you can visit it here. you should definitely check it out, and tell them i sent you. you won’t get a discount of any kind, but tell them anyway, why not. working on this was a lot of fun, and i hope i get to do many similarly odd and out-of-comfort-zone things in the future.

-selby

p.s. - here’s what louisa sent me for my testemonials section (at prohacker.co.uk, visit now to pitch me your oddly viable ideas):

“Working with Selby in the run up to my website launch made things easier and allowed me to focus on the marketing activities, leaving him to test the technical systems. He was thorough, punctual and communicated well. The testing process was facilitated by two meetings, and my personal lack of technical understanding was no barrier to Selby who was able to explain things in layman's terms so that I could be aware of the progress of the project.“

- louisa rogers, m.d. of trendlistr.com

all of which makes me sound very proper and professional, which i appreciate, it definitely lends the rest of the blog post some credibility. catch you all next time. ;)

[prohackeruk-blog]

hackercopter update

a hackercopter comes in three parts: the “hacker” part, the onboard computer which actually does the hacking, the “copter” part, the quadcopter which flies the “hacker” part into physically inaccessible areas containing target networks, and the conundrum of powering the whole thing. today, i’ve been working on the “copter” part.

the first thing i had to do was determine how much weight my drone can carry. enter this setup:

side note: i’m announing my newest product, the drone harness, for the low, low, low, low price of eight rubber bands. available in any colour(s) that you can find rubber bands in. drone not included.

now that that joke is over, this was the plan: get the drone in the air, then keep adding those black, circular weights until it won’t reasonably fly any more. the weights way 100 grams each.

this plan was a gigantic failure. that’s not to say it was a bad plan; you can’t judge an idea by its results alone. however, the drone i have isn’t capable of flying with a single 100g weight, which means i have a weight limit of somewhere between 0 and 100 grams. it came pretty close, though, so i’m hoping i have more than 50g to work with. that’s still pretty tight, but it’s probably fine for what is just a prototype.

-selby

[prohackeruk-blog]

identity crisis

above: picture not really related, but off the record, wat kitchen make really good noodle boxen. no, they aren’t paying me to say that.

so i had an idea about a problem that’s been bothering me for a while now: making web apps. more importantly, making good web apps in a reasonable amount of time. it’s all very well and good knowing exactly what you want (or what your client wants) right now, but how clear in your mind/notes is that going to be once you’ve spent a week making sure that users can log in and out correctly?

enter stage left: “the component model”. after working with a certain open-source identity service (which shall remain nameless to protect the guilty) recently, i was despairing for all of the developers who have to rely on such things, either because they don’t have time to write their own, or because they’ve inherited a project which already uses one. (a bad one. no, i’m not bitter, why do you ask. 0_0) once i’d stopped crying, i made the executive decision never to let this happen to my projects, and so went about solving the problem. that is what i’m supposed to be good at, after all.

so, starting a few days ago at time of writing, i’ve begun work on some standard (almost) plug-and-play components for use in web apps. the first, fittingly, is an identity service, complete with all of the common and useful validation checks, features, etc. etc. etc.

problem solved. #winning

-selby

[prohackeruk-blog]

intelligence is cheaper than you th_nk

so this month, like most months these days, i went to rise & design at the northern design centre in gateshead. for those unfamiliar, it’s a lot like a networking event, except it’s run by and for designers, so it’s fun. (if you’re interested, then check them out here: http://www.designnetworknorth.org/events/). come say hi, i’ll give you one of my sexy new business cards, we’ll get some bacon sandwiches. it’ll be fun.

anyway, this month’s event was about artificial intelligence. not general ai, the kind that will eventually make humanity obsolete, that’s still a while off yet -- this one was more about the integrated types of ai we already see in the modern world. three gentlemen from th_nk (read: think), a digital agency operating in the area, came in and gave talks about different applications of lesser ai. there was a discussion about the efficiency of “conversational interfaces” (talking to machines in natural languages as opposed to typing and clicking), a small demo of the amazon alexa as the world’s most intelligent shopping list, and, most importantly, mocking the google home device for looking like a damned plant pot. it’s had that one coming for a while.

sentient plant pots aside, there was one thing i found incredibly interesting. as it happens, there are a number of services which offer certain ai features as a service. natural language processing, for example. the advantage of this is the exact same advantage an api or interface usualy gives -- you have access to a feature you otherwise wouldn’t without having to do nearly as much work as setting it up your self. given that even simple functions of ai and machine learning are what the professionals refer to as “bloody difficult”, this is an incredibly valuable thing. i’m going to check out some of the sources referenced by the guys from th_nk in their talk and give a run-down of each of them, but that’s getting a post all of it’s own. until then... don’t let the machines take over. your plant pots may be more intelligent than you realise.

-selby

[prohackeruk-blog]

these things are all related. stay posted for more updates... plus some sweet drone shots once i can get the thing in the air.

-selby