Video Of The Week: Maria Klawe and Hadi Partovi at the Recode Conference

#hacking education

Comments (Archived):

  1. jason wright

    i like the ribbon of light effect at the beginning and the end of the video. a Pando rip off, or ripped off by Pando?In England in the nineteenth century the economy and population grew so rapidly that the Victorian elite decided that the masses needed functional reading and writing skills for the new tasks of the industrial age. isn’t the coding skills debate just a repeat of this?

  2. Matt A. Myers

    I wonder when we’ll include a curriculum of learning about the body-heart-mind connection in schools. Every year I learn skills that would have been good and beneficial to have known a decade or two ago.

    1. Richard

      Hearing this women essentially say “don’t do what you love, study CS, sent a shock throughout my body. Hey you girls studying Design, Mathmatics, Statistics , Elightenment, Synthetic Biology, Nanotechnology, Earth Science, Food Chemistry, Finance, Fashion, Film, and Biotechnology, medicine, global health…Stop what you are doing….

      1. Matt A. Myers

        I think balance and having multiple things someone works towards, especially something that is challenging, is important. If anything the more exposure you get to different fields, and the deeper that exposure, then the more you can interact with others and the better understanding of the world you can have. There are a lot of things I would like to learn though our societies aren’t really structured to allow me to do that easily, and the current structure of our school system certainly is generally rigid at the moment – especially in terms of time constraints trying to have people know X knowledge by Y time. And we clearly know that this current structure leads to too many people with degrees and knowledge that doesn’t give them work relating to what they studied – which could be viewed as an inefficient use of time/resources. In our current educational structure, where time is a constraint, all of those categories/life interests you mentioned will be competing for one’s time and attention. I think happiness is the better leading metric to follow though, so if someone wants to learn about Design, Mathematics, Fashion, Finance, Health, Food, and CS all at the same time – they should be able to, and that would allow for optimal happiness to be found – and they are likely to find one topic they enjoy more. Yes, we should likely incentivize people or find learning structures and material that incorporate many of these — e.g. projects that require someone to learn some math, finance … maybe some CS and design – the structure of incubators fits well for this: want to do a fashion-related idea? You’ll quickly figure out what you need to learn from the obstacles and challenges you run into – and then hopefully have the guidance to show you what the possibilities and likely options are.

    2. Alex Wolf

      Matt – agreed! My co-founder Vijal, an osteopathic doctor, could not agree more. Add to that he’s in Psychiatry, and mental health play a big role, as well as bridging the body-mind related things which arise.

    3. LE

      Super important point. Understanding of people is definitely something lacking in education from my past experience (may have changed).Might be discussed or buried in the fine print of psychology but need to be front and center.I was educated in this by family growing up, have refined my understanding of it over the years, and it’s pretty much on the top of the list of things that I use every single day (along with typing the other thing that everyone should know how to do).That said some people (example people with autism or aspergers) are going to have a really hard time with this or similar subjects.

      1. Matt A. Myers

        There will be edge cases with any topic where the person doesn’t have the same capacity/speed of learning, though like if you give enough time to people to understand mathematics and not just memorize tricks to think your way to a solution, then they can slowly learn the foundation.When you remove the constraint of time and the funnelling of people through certain specific course work/topics – pressures due to trying to optimize for productivity for GDP without taking into account much else – then the quality and nuanced learning will be deep, real and diverse.This style of learning could even include workshops for health-related and active healing – preventative and proactive healthcare – yoga, food knowledge and preparation, relationships and communication, etc.. In my mind it really just looks like a community being engaged.

  3. Richard

    Women make up 50% of PhDs in science and engineering. All of these women code in some way (many are using R, Matlab or SAS and more recently python). Many take electives from either the Stats or CS departments. They are doing what they love and are computer proficient.There are a lot of “soft bariers” to entry for young women in the startup tech space, particularly for younger women on tight budgets. For example, let’s say there is a start up event in a neighboring city, while a guy can simply crash at a seedy hotel or some random airbnb space, a girl has to think about safety and security at a different level.

    1. jason wright

      in another recent post on here i commented on the seedy location of my local hack space. what woman would dare venture alone down such streets on a dark evening in winter? even i would be looking over my shoulder.

      1. Matt Zagaja

        Maybe the women are right, and that’s why they are less likely to be victims of violent crime by strangers then men: http://www.bjs.gov/content/….Our gender may “fear” crime less but we’re more likely to be victims of it.

        1. Richard

          It’s not numbers, its impact. No comparison when rape is considered.

        2. jason wright

          the organisers of the space are stereotypical white male geeks, and they are not entirely ‘open’ in their outlook (they seem to have a paranoia that an outside group of people will sign up with the intention of taking over the organisation via the democratic constitutional process), and overlook the factors that work against a diverse membership (location being one ‘obvious’ factor).i might set up a different kind of space, a hybrid of old school arts and crafts meets techno maker modernity. we shall see.

          1. Matt Zagaja

            Sad to say having gone to a tech school that you are right that there are many geeks that are not great at being inclusive. Location is a tough issue to resolve, we’ve been discussing it with our local Civic Hack Group. The start-up that sponsors our events generously lends us its office in New Haven and pays for pizza, soft drinks, and adult beverages for our events. Finding something comparable in the suburbs is difficult, and many of the core members of the group bicycle to the event from their jobs in the city so there would be extra coordination/effort involved in getting them to a suburban event. Maybe moot if there is sufficient interest from those outside, but it becomes the classic chicken and egg problem.

  4. Alex Wolf

    I’m happy to finally have time to watch, and comment, on a Saturday! A great set of topics for it too – coding, STEM, education, and women in tech.The finer distinction between STEM and coding jobs interests me, but I find the lack of STEAM troubling. The STEM to STEAM movement, partially headed by my alma mater RISD, and John Maeda (who jumped ship from RISD presidency to Kleiner Perkins this winter). STEAM makes a huge amount of sense yet still only a certain group of people see how adding the “A” (for Art and design) is so vital to visualizing the STEM fields, making them approachable to kids mentioned in the ed sense of this talk – not just high school, but elementary and middle schools, ie. not just the kids graduating to “fill jobs” as pointed out, but to make kids native to the STEM continuum going forward.Having just released a STEAM ed app, ferret (free for iPad – http://bit.ly/Fer512 ), and having just won best digital – mobile for it a week later, I am proud of that recognition because it’s the first app I designed, because I’m a female founder, and because I’m coming from an art and design background. It’s a win-win to up numbers for women in tech and women founders while also promoting STEAM and Edtech for kids.Interestingly, the lead is sort of buried on what the app really does – it’s the first biological “periodic table” created to classify organisms. We have not made much noise about this yet since the paid version is a few days shy of release. Most ed apps simply re-platform existing content – we actually demystify and reorganize content using design, and hear great things back on how this system actually helps people understand the content better. We are coding the bio into small blocks to make them more accessible, quickly so you can push deeper easily. It’s like LittleBits for biology.Tools like this are highly complimentary to coding thinking, which along with CSNYC and the host of other awesome efforts, will make STEM as easy as language. I wish deeply for the visual side of learning to accelerate along with the coding and STEM since we live in such a visual world. STEAM and Biomimicry will exponentially push this momentum.

    1. LE

      That bitly link doesn’t work.

      1. Alex Wolf

        Nice call LE! It was the parenthesis d’oh! Should be fixed now.

        1. LE

          I will have the kids check it out. I’ve always wondered why learning has not been wrapped into the addictive games that kids play. I guess because the game developers don’t have a horse in the education race. I mean it seems so obvious.Anything addictive can be used for learning purposes almost effortlessly. Commenting on AVC has helped me in many non obvious ways for sure.I just realized that if you had asked a question “why doesn’t this link work” I would have knowing immediately why it didn’t. [1] But since I only clicked on the link I didn’t look close enough to see the “)” which I would have realized was the issue.[1] One of the fun things about coding is spotting exactly this type of thing after an error, fixing it and moving on.

          1. Alex Wolf

            Smart should be the new addiction. Give someone a key to their own intelligence and ways to grow it and they will flourish. I had a lot of push back on the design and how “unkiddy” it was, and flat. iOS7 came out and I looked like a genius.Interested to hear your and kids reactions.Have not had the ) error before since I usually command-K links, but couldn’t do that for some reason…

    2. SubstrateUndertow

      it’s the first biological “periodic table” created to classify organismsTools like this are highly complimentary to coding thinking,. . . . . will make STEM as easy as language.I wish deeply for the visual side of learning to accelerate along with the coding and STEM since we live in such a visual world.STEAM and Biomimicry will exponentially push this momentum.WOW!that really zooms out and abstracts the concept of “CODING” by taking it up the stack into the future where it can facilitate more diverse mass-culture participation.Hard Syntactic-Coding down in the STEM engine room is foundational to enabling the emergence of a larger more abstracted virtually/visually code-able tool-set-world for the rest of us.In the long run only the gifted few need to master those low-level engine-room CODING-SKILLS. There mission is to commoditize a set of higher-level reality-coding tools/narratives/metaphors to drive our coming GERERALLY-CODE-ABLE mass culture.BUTWe all need to STEAM ahead at learning to visualize our world as recode-able, recombinant, building blocks.Edit:Especially up voted for mentioning Biomimicry our much under appreciated organic cheat-sheet.

      1. Matt A. Myers

        I think I understood what you said. 🙂

      2. Alex Wolf

        It’s all about developing pattern recognition – visual and conceptual,which then output as some form linguistics, whose patterns make study easier and enable communication thereof. Pattern recognition is a top skill in survival and hence evolutionarily very valuable.You may relate it to Syntactic Coding, which if I understand correctly allows for grammar and syntax to be applied to code with abstraction. Our flip side game to ferret is Ani-gram-it ( http://bit.ly/1oVRIF7 ), which we liken to scrabble for animals – instead of spelling words with letters you make animals with body parts, “anatomical spelling”. Again all patterns – insects are hexapods, mammals tetrapods, it’s natures math where you can learn tricks like how many legs you have decides if you are vertebrate or not, warm blooded or not.Complexity theory is based on simple parts, knowing how those parts mate is about patterns.Biomimicry is about not wasting time reinventing the wheel – look to nature for solutions since she’s been designing for 3.8 billion years, and the Biomimicry Institute led by Janine Benyus is all about merging biology and design. Amen.

  5. pointsnfigures

    I am not sure that I agree with forcing everyone to do STEM. Each person is different. I have one daughter that just didn’t get math at all-wound up majoring in history. However, it was rigorous-almost like applying the scientific method to non science.I am all for more STEM study-but I think we need to integrate it into everything. Coding should become like writing.

    1. Alex Wolf

      In the world I’ve been imagining something like that exists, and the parallels in each field are highlighted, making it easy to see the structures in disciplines, and their unique divergence. The app we built is a platform for many subjects – even wines, as @awaldstein agrees.

      1. SubstrateUndertow

        the parallels in each field are highlighted, making it easy to see the structures in disciplinesThus, elucidating the fact that code-ability is so endemic to the every fabric of reality!In the abstract every field of study focuses on some platform of recombinant rules for recoding base components !

    2. LE

      I am not sure that I agree with forcing everyone to do STEM.Well of course everyone should not be forced to do STEM. Exposed to it to see if any interest, yes. Forced to learn and take course (other than perhaps an intro) possibly.My 2 daughters in no way would be good with STEM.My stepdaughter though, yes. She is doing all the robotic stuff and will come with me to the basement to help with the HVAC. (And she is pretty good at sinking baskets for that matter I just watched her yesterday and was really impressed (and that’s me talking here)). Helped build the wood board with me for the robotics competition. Like the drilling part and all of that. My stepson (sports guy) no interest in anything mechanical at all. All sports. All the time. Not impressed with the RC Helicopter. Girl is, wanted to fly it.Coding should become like writing.I have spent plenty of time with all sorts of “normals” and computers. I can assure you that there is simply a large portion of the population that doesn’t get computers, how they work, and probably never will. If you are in the business these are the people who laugh about how stupid they are with computers and think it’s funny or something. You’d have an easier time getting me on board with sports appreciation because it’s something that is an art form in a way, involves strategy and has interpersonal relations that creates drama. (And this was all before Donald Sterling..). Because my brain is wired to enjoy things like that [1] you just have to spend the time to show me how sports fits into how my brain really works. But some people’s brains aren’t wired to think they need to code.I’ll make a controversial statement here (so watch out). One reason (as far as women at least vs. men) is that theories exists that men are problem solvers and women are not as much problem solvers. (No links to give here but google it. [2]) Being able to code helps you solve problems (even allows you to invent problems to solve). This seems like something that is more applicable to a man’s brain than a womens brain from my experience. As a generality of course. Which is separate from saying women won’t or can’t be good coders.[1] I love many reality tv shows for example. And I love gossip.[2] Ok here you go: http://www.healthguidance.o

      1. Drew Meyers

        “I can assure you that there is simply a large portion of the population that doesn’t get computers, how they work, and probably never will.”The same can be said for writing. Some people will never “get” it..

        1. LE

          Except that writing is analog so even if you don’t get it you can still communicate in writing (think immigrants and broken english). If you don’t get computing you are always going to be frustrated and depend on others to get you out of messes.

  6. Rob Underwood

    I’m a huge proponent of programs, in-school and after-school, to teach kids how to code and making it part of the core K-12 curriculum. I’ve been active in NYC, and especially in Brooklyn, to connect schools with programs and advocate for why this is important to parents and wider communities. But I don’t think there’s any need to say coding is “way more important than American history.” First, I just disagree — I think they are both important and we shouldn’t create a false choice. But more importantly, statements like that are just the types of statements that are likely to turn off those parents and teachers who aren’t yet on board with why this is so important. Detractors of this effort – both those with legitimate concerns and those more in the conspiracy camp – will seize on statements like that an make our work just that much harder.

    1. LE

      But I don’t think there’s any need to say coding is “way more important than American history.” First, I just disagree — I think they are both important and we shouldn’t create a false choiceIf you only have a dollar to spend you are going to have to decide what the best use of the dollar is. Likewise with time.I think part of the issue is simply that school curriculum as far as the “must subjects” hasn’t really changed that much over the years. Things that made sense in the 30’s don’t necessarily make sense today. Exposing people to things if fine but making them spend time and testing them (to the degree of getting graded) may be overkill.And for the most part, even with people’s ability to learn in other ways because of the internet (and instants answers to questions) the way teaching is done hasn’t changed as much with the times as you think it would. At least in any typical suburban school that I am familiar with.As far as what is important or what is not people always try to justify why something should be done without considering that time is a limited resource. You can always justify anything if you don’t look at the downside (time or perhaps expense).

      1. Rob Underwood

        I get the time in the day argument but also believe it’s possible to do both at the same time. For example, when a student writes an essay in US History class a student is developing both skills in US History and English (specifically writing). In a similar way, coding can be integrated into US History in a similar way, especially given the plethora of public data sets now available. For example, a data set on causes of death, regiment, age, and state of soldier for the Civil War could be evaluated via Python or R to examine potential correlation patterns.Moreover there are lots of lost moments in the average school day in terms of transitions from topics, transition between classes, etc. that if reclaimed would also provide more time for instruction.I’m reacting especially to the phrase “way more.” If she simply said, “I’d give the edge to coding over history” maybe I’d not be reacting so much. But so many of our issues in the US, in my view, are associated with “low information voters” – i.e., citizenry on the right, left, and in the middle who have a very limited understanding of civics and history (as well as science, etc.) The implication that we could tolerate less emphasis on US (and World) History is worrying to me.Finally, as Fred knows, I’m very much on the front lines not only of education but also of talking to communities throughout Brooklyn about the need for coding in schools. I talk to parents from all sorts of backgrounds, some of whom have no understanding of what “coding” even is. Just two days ago I was at a Community Board meeting speaking with a CEC member (our legacy school boards) who lives in a large, relatively disadvantaged, Hispanic community with many immigrants. She was sharing some of the skepticism that her community has about coding programs because for some computer systems are associated with tools to evaluate documentation status. More generally I hear parents and families tell me that while they think coding is important they worry first about their children getting the “3 Rs” and even just being safe going to and from school. These concerns are not insurmountable issues and we ALL should, just as Fred is doing, advocate for ALL children to get access to coding instruction. I just don’t think saying coding is “way more important than American History” would make a very good opening point, and in fact makes our case harder to make.

        1. LE

          For example, a data set on causes of death, regiment, age, and state of soldier for the Civil War could be evaluated via Python or R to examine potential correlation patterns.Great point. Similar to how I have learned to write programs. By having a real problem that I need to solve as opposed to just learning concepts out of context (which makes it much harder to understand and digest).Moreover there are lots of lost moments in the average school day in terms of transitions from topics, transition between classes, etc. that if reclaimed would also provide more time for instruction.This would be quite difficult given the mediocrity of how most people operate. It would take a true “jobsian” push toward excellence with legacy getting trampled. Given unions, entrenched administrators and complacency I’m not sure how achievable it is on a widespread basis. Of course if you can tie some reward system into making the change (and design the reward system to make it less likely to game and have unintended consequences) it is a possibility for sure.Personally I think it would be a good idea, like the guards at the german prisons [1] to rotate teachers and not have them stuck to the same school forever and ever. And if you have tenure (whole separate issue) develop a system to make it transferable between school districts.

          1. Rob Underwood

            “a true ‘jobsian’ push toward excellence (in teacher quality)” == day job.

          2. LE

            By “jobsian” I mean Steve Jobs.

        2. sigmaalgebra

          For some of how American history has been taught too often, sadly coding might be more important!Usually telling much of the truth about history is so politically incorrect that ‘it just isn’t done’. Also, history as a ‘discipline’ is a bit short on objectivity, identifying causality, and ability to make predictions that are meaningful and accurate. Still, history would’a, could’a, should’a be a fascinating subject and of high importance for the health of our political system.

      2. Richard

        There is no single economic utility function, the economic utility of the ith person doens’t equal the economic utiity of jth person. I certainly believe that programming literacy should be required for highschool graduates (even at the most basic “hello world” level). But, education is not vocation. Lets not forget the what makes America work. In a word, Freedom and Capital. For it is Freedom and Capital that gives us the Gordon Moores.

  7. Frank W. Miller

    So, I’m trying to figure this out. I hear from many tech leaders that “we need more women” in CS or coding or whatever. They then launch into their favorite method supposedly increasing those numbers. But I’m still back at the why is it important for us to make sure that there are more women in this field? I think it would be great if there were more women but I don’t get why its important to try to further that. It just seems like its too much of a PC thing to me My 30 years of experience in computers yields the observation that women just don’t like to code as much as men on the average. Maybe computers are just one of those fields that women aren’t interested in?

    1. vruz

      Let’s see if I can understand you…You only want to know what’s in it for you, and you’re wondering why women don’t want to work with you?Hint: maybe it’s not the code.

      1. Frank W. Miller

        You completely misunderstand me. Not sure how you an leap from wondering why its important to increase the percentage of women to what’s in it for me. The question still stands. There are other fields that are heavily biased towards one gender or another, the NFL, nursing, construction work, administrative assistants, etc., etc. Maybe this is just a field where one gender has a preference for it and the other does not. What’s the big deal?

        1. Matt Zagaja

          Assuming no inherent differences between the quality of male and female developers (and I don’t think there is any reason to believe there is) if you’re the one doing the hiring and women aren’t coming in through the pipeline you’re basically getting 50% of the number of potential applicants you would otherwise be getting to choose from.

  8. Twain Twain

    The inclusion of female engineers is VITAL to the intelligence and the innovation of systems. The earlier it becomes a compulsory subject, the more it sends a signal to girls that it’s a norm just like maths and language.It may not be obvious but the reasons why AI is autistic (which IBM Watson’s inventor called it and which Google’s Star Trek computer is, by design) and has not solved its Natural Language problems etc. is because intelligence and language is emotional as much as logical, rational and probabilistic.Emotions are a domain expertise of female engineers by virtue of the neuroscience showing that our hippocampus is 11% more adept for memory, language and emotions. This has implications for AI, memory chips etc designed by female thinking and crafting.Google this week released its diversity metrics so I’ve invited Larry Page and other SV tech executives to join and sponsor “Mission to Mars: eCommerce and AI built by women not bots”.Etsy, Pinterest and Square have been cited as examples of best practice in fostering pipelines of female engineering interns:* https://plus.google.com/113

    1. Twain Twain

      Comp Sci was compulsory in my school from 11-14. My Comp Sci teacher was a woman as was the Head of Maths and my chemistry teacher. My mother was an engineer so I’ve always had technical female role models.I’d gotten into tech when I was 8 after my Dad gave me a games console and a light saber. We had the Star Trek Enterprise too but I really wanted a Millennium Falcon!My grandfather had discovered I have a mathematical brain when I was 2 so whilst my parents let me wear my pink gingham dresses, sparkly shoes and be a “girlie girl” because I loved that, they also encouraged me to get into the sciences and be technical.My mother taught me to count via card games not with my fingers. My Dad taught me to play chess and about circuit boards, catapults and carburetors.Girls can be feminine whilst also into “boy” subjects and that’s a signal that education, teachers and parents can send.

    2. sigmaalgebra

      > the reasons why … is because intelligence and language is emotional as much as logical, rational and probabilistic.I can agree with the part after “because” but not the part before.When I worked in AI at Watson, I didn’t see any hint of even a weak little tiny clue of promise of AI. We were trying to apply AI to monitoring and management of server farms and networks. Okay? That’s the problem? Right? Okay. Good. So, to heck with the AI nonsense, and let’s attack the real problem. So, I stirred up some new soup in essentially ‘applied probability’, got some nice progress on monitoring, that could ‘learn’ and ‘adapt’, and published it. It was NOT AI, but it was better progress on the real problem than all the AI stuff we did. Progress? What’s wrong with that? The AI stuff didn’t give any progress, but some applied math did. At least for now, debate over.For now, that ‘paradigm’ is ALL that’s promising. E.g., we gave a paper at an AAAI IAAI conference, and essentially all the progress at the conference was from just the traditional, careful, well considered engineering ‘paradigm’ with no real progress on AI.Traditional good work can give lots of successes, but we dance ’round and ’round and suppose while real AI is a secret that sits in the middle and knows and we don’t.My view for “because” is simpler: We don’t have even as much as a weak little hollow hint of a tiny clue about how real intelligence works or how to write software for AI. There are and will be many opportunities for software to do amazing things that in the past needed an intelligent human, say, playing chess, but a chess program is hardly even a start on competition with human intelligence.I have some vague thoughts for AI that would make a very challenging project, likely not promising, but I also have very precise thoughts and a very promising and entirely doable project I want to get done. So, AI goes to the back burner, and that’s likely where it belongs.

      1. Twain Twain

        Ah and have you seen my slides on:(1.) What philosophy of the mind and Neuroscience can teach the machines about learning?* https://plus.google.com/113…[Yes, you’ll see a nice image of IBM Watson’s autistic architecture as well as Google Star Trek’s autistic architecture.](2.) Information intelligence: human, business, machine and quantum?* https://plus.google.com/113…Chris Dixon of Andreessen Horowitz’s advice to founders was:* You need to know a secret;* Know the tools better than anyone else;* Know the problems better than anyone else; and* Draw from your unique life experiences.Furthermore, he said that these are the characteristics of innovation:* Powerful people dismiss them as toys.* They (the innovation) unbundle the functions done by others.* Does it challenge social norms?Well, I know the secret for why AI is essentially not intelligent. It has its stem (pun intended) in global society separating our emotions from our logic circa C18th French philosophy.Thereafter, during Alfred Binet’s 1905 invention of the IQ test which is now basis of GMAT etc, intelligence became to be defined as logic (that’s all those series and sequences questions), rationality, geospatial reasoning and pattern recognition of “odd one out wards.It’s only in the last few years with advances in Neuroscience that we’re empirically re-examining the factor of emotions in our language, memory and intelligence.That challenges existing social norms of intelligence being pure logic, rationality, probability.

        1. sigmaalgebra

          > Mathematics are expressions of metaphysical objects.About 100 years ago, there was a lot of work on the ‘foundations’ of math.After about 70 years, the picture became clear: If believe in elements and sets of elements, then with some cute side considerations, e.g., to rule out Russell’s paradox, get Zermalo-Fraenkel set theory and, there, can build all of math on sets. So, if you are ‘comfortable’ with sets, then you might also be comfortable with the rest of math.Key results include those of Kurt Gödel and Paul Cohen.The current field of model theory tries to say more.I can’t take Dixon’s ideas very seriously. Besides, no VC firm would invest in anything based on the points you gave. For one, they would refuse even to evaluate such points. For another, they believe in essentially a Markov assumption: Such points and the future of the business are conditionally independent given current traction. So, they consider traction and f’get about any such points.> Well, I know the secret for why AI is essentially not intelligent.There is nearly no end of ways to be wrong; AI has found lots of them.Binet’s work on ‘IQ’ shows yet again how easy it is to do sloppy work.> That challenges existing social norms of intelligence being pure logic, rationality, probability.Sure; yet again it’s easy to be wrong, so many ways to be wrong.

          1. Twain Twain

            Yes, this is true. No VC would invest in any founding team that went in and said:* We know a secret;* We know the tools better than anyone else;* We know the problems better than anyone else; and* We’re drawing from our unique experiences.VCs of today are not like the patrons of centuries past like the Medici family who commissioned giant projects to be created from scratch even before there was any proof it could technically be done.Today everything is about pre-existing proof of concept and traction before VCs invest.And it’s about quantifying that traction.

          2. sigmaalgebra

            There is for projects I would consider serious a well known, high quality solution: Present the project plans just on paper. Have the paper reviewed and evaluated. Then decide to fund the project or not.Does it work? Heck yes, and has a much better track record than VC funded projects. Examples? Sure, nearly everything funded by NSF, NIH, and DARPA and a huge fraction of everything else funded by the US DoD. Also, many projects in engineering, e.g., tall buildings, long bridges, new submarines, new airplanes, etc.Why don’t VCs do this? (1) Nearly no VCs have the technical backgrounds of problem sponsors at NSF, NIH, DARPA, ONR, USAF Cambridge, US Army Durham, etc. or the staffs at Boeing, General Dynamics, Lockheed, Raytheon, etc. So, the VCs can’t review the plans and, really, would have a tough time even directing such reviews. (2) The VCs likely don’t get many such plans; the people who could write such plans quickly learn that VCs won’t read what is written and do something else. So, VCs get something on 100 new takeoffs on SnapChat or some such. (3) Likely the LPs want their VCs to be more ‘conservative’ and evaluate projects more like PE firms or even commercial bankers.

          3. Twain Twain

            I used to do Strategic Investments (TMT) for UBS so have some insights into LP and VC approaches. After that, I was employee #4 at a merchant bank start-up; some former Heads of banking at various institutions decided to try and create their own merchant bank so they hired me.One of the focus areas was Renewable Technologies.Since most of my colleagues had studied liberal arts or gone to business school, their frames of reference for science was limited. I had gone to business school but had a background in science.So I was the one despatched to the European Wind Energy Conference to assess and compare the engine turbines and I was the one who read the technical documents on redox batteries.Prior to startups and banking, I’d spent my teenage years working in the chemical industry in product development so I have some frames of reference about how different the financing process is in industry vis-a-vis VCs in social media startups.

          4. sigmaalgebra

            The situation was dumber than paint but crucial: In Visual Basic .NET Framework 4.0, object instance de/serialization (that is, convert the instance to an array of bytes for storage on disk, in database, transmission to another computer, etc.) was failing: Worked fine when did both the serialization and deserialization in the same program, but the deserialization failed, with an ‘exception’ raised by the relevant (‘binary formatter’) Microsoft software, after transmission via TCP/IP to another program (another server in my ‘server farm’).Guess: The objects, instances, and their serialization have ‘long names’, longer than the programmer easily sees, and the deserialization wants to go only to an instance of the same class with the same long name as the serialization, or some such. Well, then, basically have to declare the class in a file that can be used both by the software doing the serialization and the software doing the deserialization. So, for this file, try to use a ‘dynamic link library’ (DLL) or ‘assembly’.So, tried a DLL, the one my Web pages are using successfully for the serialization. Blow up, from Microsoft’s software. Bummer! How the heck to get my server trying to do the deserialization to make use of the same DLL? The best documentation is clear as mud. So, we’re talking super @JLM barbed wire enema, self-inflicted, unanesthetized root canal procedure.Solution: Consider all hints and possibilities and then all combinations and try them all, keeping records. I.e., the usual way: Throw it against the wall different ways until it sticks.Result: As of last night, it worked! No blow up!Then the server software stopped with an error back from a routine of mine for a fancy version of binary search, but that’s MY code and easy to debug. And the code has already run well in some fairly good testing. So put in some trace statements, run again, find where the problem is, and fix it, piece of cake. So, it’s back to my code and not just mud wrestling with bad documentation and throwing stuff until it sticks. Not a good foundation for good engineering, but that’s the way of that part of the world for now.The mud wrestling, @JLM barbed wire treatment, root canal procedure took far too long. Such problems from bad documentation have been the only difficult part. There were several ways around the problem, but serialization using a binary formatter should be nice, fast, only a little code, flexible, if it can be made to work.Old message: ‘Ease of use’ is from tools that are well designed, reliable, and well documented and not from tools that try to do everything for you. E.g., it’s much easier to chop onions with a good cutting board and chef’s knife than with an onion chopper; it’s much easier to cook some hot dogs in a pot with some water than in a hot dog cooker; it’s much easier to write software with a good text editor and some command line scripts than with an ‘integrated develop environment’ (IDE).

      2. Twain Twain

        As for chess, when I was 5 I had no construct of probability which is said to govern chess rules and outcomes.Yet I beat my father in my second game and consistently thereafter.The assumption has been that chess is a game of probability.What if it’s not?What if it’s a game of perceptions?Since I had no construct for probability, my child brain was not computing the % likelihood of win/lose risk of each move.I navigated by perceptions, geospatial reasoning and the rules governing where pieces could move and how capture happens.Those perceptions embodied my emotional state and my interpretation of my Dad’s emotional state. If he made an aggressive move, it wasn’t just because he wanted to intimidate me it also meant he was trying to distract my focus away from a weakness in his defense.That’s the thing about machine intelligence — even chess playing ones. They’ve abstracted the emotion factor that informs decision-making.Yet the ancient Greeks and modern Neuroscience direct us to a renaissance of emotions in the way we model intelligence for the human and the machine mind.

        1. sigmaalgebra

          For what you are saying about probability and games, chess is not the best example of a game because in basic game theory, e.g.,T. Parthasarathy and T. E. S. Raghavan, ‘Some Topics in Two-Person Games’.chess is a game of ‘perfect information’ and in principle there must be optimal play where no probability need be involved. That is, for chess, there is just a tree, and all we need to do is enumerate the tree and see how to play. The current practical problem is that so far the tree is too large for any direct enumeration. So, we can try various techniques to do the same but faster; still we don’t know how to play perfect chess.> As for chess, when I was 5 I had no construct of probability which is said to govern chess rules and outcomes.Again, in principle there need be no role for probability in chess — the game is too ‘simple’ for that; and that is why chess is not such a good example for your role of probability in games.But, if take an advanced approach to probability (Kolmogorov, Loeve, Neveu, Breiman, etc.) or just swallow some things, then each move of your father is a random variable and will have a distribution from which in principle one could calculate probabilities, and those could be an aid to winning if don’t (and so far we can’t) actually play an optimal game.> The assumption has been that chess is a game of probability.Well I’ve tried to outline that, yes, one could use probability but, still, the game is not necessarily one of probability because there has to be optimal play that has nothing to do with probability.> What if it’s not?It’s not. But for optimal play, where there is no use of probability, so far that’s too difficult. So we are left with various approximations, heuristics, etc.: Say, have a computer look 20 moves ahead; although that still stands to lose to optimal play, in practice against humans it might pick up some grand master points.> What if it’s a game of perceptions?For optimal play, it can’t be; that is, with optimal play, which we already know has to exist, it doesn’t matter even a weak little hoot what any ‘perceptions’ are; however, in practice, with human players, maybe sometimes it is a game of perceptions.> Since I had no construct for probability, my child brain was not computing the % likelihood of win/lose risk of each move.Actually “computing” probabilities is much more often imagined than done; in essentially any real situation, with just astoundingly meager assumptions (so meager it’s tough even to think of anything that violates them), we know that probabilities have to exist, but we rarely have enough information to say what the probabilities are (e.g., flipping a fair coin) or enough data to make very accurate estimates.Thus, about all humans can do to ‘compute’ probabilities is to apply some intuition about what might happen on, likely, a fairly simple scale of, say, essentially always, very likely, somewhat likely, a coin flip, somewhat unlikely, etc., and here humans are making just guesses.But my kitty cats do such all the time: They are cautious little animals and don’t want to get in front of street traffic, to be at risk from a dog, to jump and fall, etc. E.g., when I went for a few dishes of Chinese food carry out two days ago, when I got back my little boy kitty cat knew I was gone, knew I’d left in the car, knew I’d return in the car via the driveway, and was waiting for me on the grass on the street end of the driveway and off both the street and the driveway. Smart little guy.My kitty cats know that they are fully safe from nearly everything in the house but hesitate at the door and take a first step out with great care. “There are old pilots and there are bold pilots, but there are no old, bold pilots.”, and nearly all mammals and likely others know and use this prudence.> I navigated by perceptions, geospatial reasoning and the rules governing where pieces could move and how capture happens.What you were doing had essentially nothing to do with optimal play, but so far that is true for nearly everything we can do with chess because the tree is too large for anything else. So, your play was a ‘heuristic’. Okay. It can be super tough to analyze the effectiveness of a heuristic; the usual way is just to play a lot of games and keep score. On some average, some heuristics are better than others; again, it’s tough to analyze heuristics.> Those perceptions embodied my emotional state and my interpretation of my Dad’s emotional state. If he made an aggressive move, it wasn’t just because he wanted to intimidate me it also meant he was trying to distract my focus away from a weakness in his defense.Girls are fantastic, astoundingly early, at perceiving the state, especially emotional, of others, especially their fathers. Beyond belief. In the crib, girls are paying attention to people, especially faces and emotions, and boys, to things, say, trying to hack the latch on the crib. Looking into the eyes of your father might have helped your heuristic. It seems that looking into the eyes is important in poker.> That’s the thing about machine intelligence — even chess playing ones. They’ve abstracted the emotion factor that informs decision-making.As we know well from the properties of optimal play in chess, if actually had optimal play, then could totally ignore all emotions; they just would not affect the outcome — the optimal play would win, period. So, likely chess programs don’t try to estimate the emotional state of an opponent. Of course, you are fantastic at estimating the emotional state of your father, and there’s no hope soon for a computer program to do any such thing; e.g., the computer program knows only the moves on the board, can’t see your father’s face, and, the program or programmer, wouldn’t much know what the heck to do with the images if it had them.> Yet the ancient Greeks and modern Neuroscience direct us to a renaissance of emotions in the way we model intelligence for the human and the machine mind.If can actually get some data on emotions and actually use it in some effective way, then, sure, go ahead. Two big ifs. Maybe in trading pork bellies, if see some really big orders, say, going long 100 cars, then infer that some fat cat is going wild driving up the price. So, close out short positions for a while, let the fat cat have his run, and then go short. Maybe traders do such things. Then maybe some trading software does also.For some more, you might want to touch on von Neumann’s saddle point results for two person game theory. Here he cuts out a lot of what one might guess would be important and gets a very cleanly posed problem and a very clean solution for it. The work is much simpler and much more clean than might be guessed.The math drops out nicely just from the basic theorems of linear programming, which actually is a quite powerful and versatile tool for proving theorems super tough to prove otherwise. E.g., for positive integer n, in real Euclidean space of dimension n, every linear functional bounded above on a finite intersection of closed half spaces achieves its least upper bound. Hint: The proof is not based on compactness! Instead, the linearity of the boundaries of the half spaces is crucial since easily the result is false otherwise. Using the linearity is less about classic techniques in analysis and more about algebraic and even combinatorial techniques. E.g., one of the cleanest solutions is via Bland’s rule, all combinatorial (he saw first for matroid theory!) for linear programming. Sometimes reality can be surprising and cute! BTW, this exercise is from W. Fleming, ‘Functions of Several Variables’.While linear programming theory is a rock solid way to knock off this little exercise, might find a simple way to argue that it is enough to consider only extreme points and that there only finitely many of those. Done.BTW, Nash’s famous result also falls to linear programming, e.g., with Lemke’s proof in Parthasarathy and Raghavan.

          1. Twain Twain

            I hacked the latch on my grandparents’ gate when I was 2; that’s how my grandfather discovered I might be technically-minded, haha.He was fixing the roof of their bungalow. I was supposed to be sitting on the lawn playing with my toys. He’d walked a few steps and turned his back to me to get some tools and I went AWOL.He told my mother he almost fell off the roof he was so panicked about where I’d disappeared to. He ran out into the street and couldn’t see me anywhere.So he made house-to-house enquiries with his neighbors.Several houses later, he found me playing happily with my friend.Grandfather told my mother he’d asked why I left and I said, “To go and find my friend.” He found it bewildering that I considered a 7 year-old my friend AND that I had navigated to her house without getting lost or crying.My mother likes to remind me of this personal history whenever she wants to scold me for having a free will and ideas of my own: “Even when you were two…..!”That’s maybe why my parents encouraged me to do “boy” subjects. I had great problem-solving and geospatial reasoning abilities as a toddler.

          2. Twain Twain

            Well chess is a combinatorial game. To an extent there are heuristics of adapting play according to learned experiences of the other player’s habits — e.g. if they constantly castle first or if they open with Reti.However, by perceptions I don’t only mean reading my father’s emotions and therefore his drivers to win. I mean the value he’d put onto pieces that was different from the standard values of 1 = pawn, 9 = Queen.The value was dynamic to him, according to whether he could afford to lose the piece or not in the contextual schema of which pieces he had left.That too is an intelligence which the machines built for chess don’t factor in. The subjective value of each piece as assigned by each player.For example, both the knight & rook have objective standard values of 3. Yet some players would prefer to lose the knight rather than the rook because, subjectively in their views, the bishop has a wider number of squares on the diagonal where it can move (offensively and defensively) whereas the knight only has 4 potential squares around it it san move to.Ergo bishop has a higher subjective value than the knight.That’s my example of perceptions affecting chess rather than logic, probability or even the heuristics of pattern recognition.

        1. sigmaalgebra

          I followed your URL but was not able to read the resulting Web page. The main reason was that my screen is 14″ on a diagonal, and the resolution is 1024 x 768. The Web pages of your URL are way, way too wide for my screen, and the horizontal scroll bar is turned off. If I use Firefox zoom to get all of the Web pages on my screen, then the text is far too small to read even with a magnifying glass.To read the screen, I’d have to capture the HTML, CSS, JS, etc. and ‘hack’ it.For AI, my view is that so far it’s 99 44/100% hype and the rest water.For the work that one might want to use AI to get done, often there are good ways to proceed now; those good ways are essentially just common, technology problem solving, e.g., engineering and applied math.My view is that so far no one has anything like good ideas for writing software for anything like real human, or even kitty cat, intelligence.Since “never” is a long time, I have to suspect that eventually such ideas will be created. Maybe you will have the key insights. But for now, any real progress in AI is a very long long shot.Now that I got my problems with de/serialization and DLLs solved (had to discover the tricky, essentially undocumented secrets), I’m back to the rest of the debugging of my code for my project. That work is now on the front burner.

          1. Twain Twain

            Ha! Well Google Plus team should have got you onto their Quality Assurance team before they shipped their image and slide features!”Big Data” and AI limitations explained in the FT:* http://www.ft.com/cms/s/2/2

          2. sigmaalgebra

            Okay, I read the ‘FT’ piece.Apparently the author talked to some well informed university people — relatively good.There is the famous R. Hamming quote,”The purpose of computing is insight, not numbers.”,and the article has“But nobody wants ‘data’. What they want are the answers.”.The article is correct about the difficulty of establishing causality and that correlation is not causality.Actually, under some circumstances there is a middle ground.Broadly the article is misleading because it gives a long string of intuitive nostrums pretending to provide information. No, such nostrums are NOT how the work should be done. Instead, as I’ve posted at AVC.com often, go through(1) real problem (2) reasonable assumptions from the real problem (3) a mathematical problem with the assumptions (4) a mathematical solution to the mathematical problem (with possibly quite a lot of computing to do the data manipulations specified by the mathematical solution) (5) reasonable interpretations of the mathematical solution (6) from the interpretations, the solution to the real problem.For these steps, the article omitted the role of the mathematical assumptions and the statement of the mathematical problem. Had these steps been done, most of the pitfalls listed in the article would have been obviously solved or not. The nostrums would have no role, nor would the broad warnings of the article.The article shows yet again that there are many ways to be wrong and to hype bottles of snake oil.More generally, the ‘FT’ and the rest of the popular media are not good sources of progress in applications of computing, e.g.. applied statistics, ‘big data’, ‘machine learning’, ‘AI’, etc. There are some good sources, but they have to do with, in some circles, such currently discredited institutions such as college and graduate school!

  9. LE

    Separately when is this “change the world” (attached from an email I just got from “Draper University of Heroes” dfj.com ) nonsense going to end.To many visionaries and not enough Indians. Stop telling young people they have to change the world. We don’t need more heroes. There are billions of people in the world and the heroes will emerge naturally.

    1. Matt Zagaja

      Likely when the A/B tests tell them that it isn’t working anymore.

  10. LE

    Sad story and related to the concept of pushing kids beyond their ability and most importantly their mental health limits. I seem to remember stories about things like this happening in Japan.http://nypost.com/2014/05/3

    1. dier56

      The need to humiliate says something in itself. But what else was weighing so heavy on that child.

  11. Richard

    I didn’t leave Caltech as an entrepreneur. I had no training in business; after my sophomore year of college I didn’t take any courses outside of chemistry, math, and physics. My career as an entrepreneur happened quite by accident Who is this? Just a guy named Gordon Moore.

  12. William Mougayar

    The answer to Walt’s first question is Yes. Maybe we should call this now STEMP, where P is for Programming.

    1. Alex Wolf

      or Pattern Recognition which is a root skill of Programming in my view.

      1. William Mougayar

        can you teach that?

        1. Alex Wolf

          Most definitely. And it’s one of my games core strengths.It’s why teachers love us so much, as well as scientists, for whom it’s a crucial skill. People like Jerry say much of being a good listener/observer is recognizing patterns. Neil deGrasse Tyson says it’s a top survival instinct. It’s across the board.

  13. lonnylot

    When I look around I literally can not find a single thing that was a computer was not involved in or did not have some impact on. People don’t need to learn how to code so they can get a job. They need to learn how to code so they can understand how the world around them works.

  14. 3agonists

    What would you say about Snapchat CEO Evan Spiegel’s email? What really bothers is that he doesn’t sound like a new person at all, but seems to be the same guy even today. Actually, are those young guys making this world a worse place in a sense? When those guys become the leaders in business, what would happen to this world? Going back to the days when sexists and racists dominated this world, is it what we really want? It is time to stop justifying unacceptable things.

  15. sigmaalgebra

    Klawe: “Coding … is as important as learning to write and as important as learning to understand mathematics.”No, sorry Dr. Klawe. Not even close.First, “coding” itself is easy, dirt simple. We’re talking what here? Define some storage, say, as ‘classes’. Manipulate the data in the storage with expressions and assignment statements. Then use try-catch, allocate-free, if-then-else, do-while, call-return. Got it. Not worse than break an egg, separate the egg, whip an egg white. Not worse than jack up a car, remove the hubcap, use a wrench with a good socket and a long handle to remove the lug nuts, change the tire. Nowhere nearly as hard as playing the A major scale on violin, in time and in tune. Trivial compared with playing the Bach ‘Chaconne’ on violin. Nowhere nearly as difficult as even a small part ofhttp://www.youtube.com/watc…right, the dream of half the population, their very own ‘CoppĂ©lia’ doll, with some of the prettiest girls and light music in all of the solar system!The ‘coding’ is much easier than writing clear documentation for what a program, function, or subroutine actually does and how to use it.E.g., relational database as in, say,Jeffrey D. Ullman, ‘Principles of Database Systems, Second Edition’,is fast, easy reading, say, over dinner at the Mount Kisco Diner. Actually getting ‘users’, ‘logins’, etc., straight with an actual installation of SQL Server, while conceptually simple in terms of the now classic and fundamental concepts of ‘capabilities’ and ‘access control lists’, NOT understood by the writers of the documentation, is much worse than a JLM barbed wire enema and comparable with a self-inflicted, unanesthetized root canal procedure. No joke: Once I got a copy of CorelDraw! actually to work on OS/2. The cost: From grinding teeth, cracked two molars. Got a crown on one, and later lost the other one.Current frustration: Microsoft’s Visual Basic .NET Framework 4.0 and, there, to the compiler, /reference, /imports, in the source code, Imports, and otherwise assemblies, namespaces, modules, and applications.Problem (1): Define the darned terms. Problem (2): Motivate and explain the concepts. Problem (3): Give several rock solid, actually work, EXAMPLES that, did I mention, actually WORK? Problem (4): Use the terms consistently. Problem (5): Do NOT use synonyms for the terms. Problem (6): At each first use of a term on a page or in a section, give a link back to the definition of the term. For any single violation, I have a 500 yard spool of rusty barbed wire ready to apply.These are all lessons in what? First hint: Writing. Second hint: Technical writing.I’m not pissed off. I’m way beyond pissed off.For the .NET problem, solution: Do a lot of guessing of what else might be relevant. Follow thin, vague, nearly meaningless clues as if on the trail of an insane night stalker Make a list of nearly all possible combinations of options. In an organized way, try them one at a time, keeping records.Result: As of last night, apparent success although I need some more testing. Now, Dr. Klawe, THAT is the !#$%^&*() bottleneck and NOT ‘coding’. All my ‘coding’ is just fine, thank you, fast, fun, easy, in 99 44/100% of the cases just trivial. The problem is HORRIBLY badly written documentation.In computing this situation is old: Once get something to work, KEEP IT, treasure it, guard it. The old case was, once get the job control cards to work, laminate them in plastic and hang them by their corners with a bead chain used for bathtub stoppers. Lock the chain to belt as crucial for life. No joke.I read the Kuhn-Tucker conditions, the polar decomposition, the fast Fourier transform, and power spectral estimation, understood it all clearly right away, and wrote the corresponding software, piece of cake, no hangups or bottlenecks. Did the same for nonlinear duality theory and Lagrangian relaxation. Dirt simple. SQL Server users and logins? Worse than JLM’s barbed wire.There’s no )(*&^%$ excuse for this outrageous, tooth cracking incompetence in WRITING. Dr. Klawe, did I mention WRITING? Did I mention that the bottleneck was WRITING? Why do I keep mentioning WRITING? Now can you say back to me what my concern is? Hint: Did I mention WRITING? Are we learning yet? About WRITING?Indeed, after the dirt simple stuff of if-then-else, etc., the main bottleneck in computing now is just writing in, say, English. Indeed, the situation is so severe that in hiring someone for computing I’d rather they be able to write a really clear and well organized term paper with no knowledge of computing than be a good computer science graduate. Indeed, computer science graduates have likely picked some just awful habits: (1) Use a symbol as an adjective. Outrageous. (2) Write software where the user has to do experiments to know how to use it, e.g., get a permanent link to a URL of a post by doing a right click on the time since the post was made; where did the guy who dreamed up that total obscurity get the really strong funny stuff he’d been smoking? (3) Use icons instead of words — we HAD icons and then got the Roman alphabet, and that was fantastic progress, 2000+ years ago. We’ve got some ‘rapidly moving’ technology here, right?If a job candidate also knew how to read/write math as a math major at the college junior level or above, much better still; my experience is that usually college comp sci profs are not very good at this.Beyond writing, for the future, the bottleneck for progress in anything technical in computing is flatly just mathematics. Period.For being able to develop software useful in the real world, currently the struggles are not at all with the ‘coding’, that is, if-then-else, etc., but with getting enough understanding of the needed, but often missing, content in badly written documentation for important tools, system and software installation and configuration, data backup and recovery, especially of data crucial for the operating system, security, system monitoring and management, network installation, configuration, monitoring, and management, applications software design and specification (as writing in, say, English), software testing, revision, and versioning, and management of software projects involving more than just a few people.Saying that ‘coding’ is important for practical computing is like saying that breaking eggs is important for running a Michelin three star restaurant.

    1. Twain Twain

      If coding was presented as an integration between maths, language and art that would be closer to its nature.It is the case that functional programming is “dirt simple”. It’s about frameworks, classes, variables, operations, outputs, verification testing and looping which follows mathematical processes.The language piece happens in mapping from a specifications document for the system which is written in English into functional notation — e.g., “When the user clicks the back button, it will take them to the next page” becomeswindow.onload = function () { if (typeof history.pushState === “function”) { history.pushState(“content”, null, null); window.onpopstate = function () { history.pushState(‘new content’, null, null); // Handle the back (or forward) buttons here // Will NOT handle refresh, use onbeforeunload for this. }; } else { var ignoreHashChange = true; window.onhashchange = function () { if (!ignoreHashChange) { ignoreHashChange = true; window.location.hash = Math.random(); // Detect and redirect change here // Works in older FF and IE9 // * it does mess with your hash symbol (anchor?) pound sign // delimiter on the end of the URL } else { ignoreHashChange = false; } }; }}The art element is both on the front-end for design and also in the analytics middleware and the database architecture and output (data visualization of the results).So, if coding is integrated in this way, that’s when orchestration and music happens.Coding is like being able to learn and play the notes, scales and chords in music. Technically, youngsters can be taught these skills.Of course whether some of them grow up to be maestros and von Karajan is a another matter but enabling them to be skilled in coding is a good start — for both girls and boys.

      1. sigmaalgebra

        > If coding was presented as an integration between maths, language and art that would be closer to its nature.My old description was that the programmer stands between the computer on one side and the real problem to be solved on the other. Part of his (her) job is to find a solution and, then, get a ‘software architecture’ by decomposing the solution into pieces that are relatively independent, small, efficient, easy to describe, develop, test, document, and robust to likely changes in the real problem being solved.> It is the case that functional programming is “dirt simple”. It’s about frameworks, classes, variables, operations, outputs, verification testing and looping which follows mathematical processes.I’ve avoided ‘functional’ programming; since Backus pursued that, it’s very old stuff; there might be some value there, but so far for me the bottleneck is just in documentation, systems management, etc. For me, on Microsoft, Visual Basic .NET is fine except for the documentation, systems management, etc.Yes, software is necessarily mathematically something, understood or not. So, for more progress in the ‘technical’ aspects of software, we have to expect that mathematical approaches will be important; actually mathematics is the only promising toolbox and methodology we have, and I’m not waiting for a better one.Athttp://www.youtube.com/watc…isDvorĂĄk: Symphony No. 9 “From The New World” / Karajan · Vienna PhilarmonicAthttp://vimeo.com/80715492isFranz Schmidt: Notre Dame – Intermezzo – Herbert von Karajan & Berliner Philharmonikeror Quasimodo looking at Maureen O’Hara. Yup, I understand exactly!It could be possible to make more out of the cymbal crash!Good composing from a cello player for Mahler.Can get the MP3 of this piece by Karajan, from the weak cymbal crash I’m guessing the same performance, from Amazon for $0.99.Maybe it was great fun to go to a Mahler concert in Vienna near 1900 or maybe just a dance at the Schönbrunn Palace playing, maybe, the waltz C. Weber ‘Aufforderung zum Tanz’ (Invitation to the Dance) http://www.youtube.com/watc…Ah, the poor, poor suffering people who listen to something other than such classical music!

        1. Twain Twain

          Haha, I literally posted that DvorĂĄk-Von Karajan piece on Google Plus a few weekends ago alongside some samba jazz!You do realise the programmer and computer on one side, the real-life human case on the side is the setup for the Turing test, right.LOL!No wonder we haven’t made that much progress in Machine Intelligence!

  16. Twain Twain

    Doing my bit to pragmatically solve the diversity issue in technology:* http://www.m2mars.com“Mission to Mars: eCommerce and AI built by women not bots” will have a Code Challenge and Talent Showcase in mid-Oct in SF.We are not about raising social awareness and discourse re. lack of women in tech; in some ways that message is a broken record. Confucius did say, “When it’s obvious that the goals cannot be reached, don’t adjust the goals, adjust the action steps.”So M2Mars is a platform for showing & telling how women contribute bottom line success to technology and innovation.