2,601 karma · joined April 8, 2008
This is especially useful if you can’t find a driver that will take you all the way.
TL;DR: The extreme customization of X (window system) is essential, "one size does not fit all", WIMP (windows, icons, menus, pointing) are not the future, audio is.
If anything, we've seen the opposite: consistency wins, consistently. It's Apple's core strategy (remember the "gasp" around the idea phone Apps have to be approved?). Google spends enormous effort on promoting consistent design in its open ecosystem.
In other words: it's less important that an upper-right "X" means "dismiss/close", and it's more important that it's consistent. (At least if you want mainstream use).
Otherwise, "audio is becoming more common", is slowly happening. And, "there should be little or no difference between local ... and Web" -- absolutely!
How would this work when the package manager can be run by anyone?
2FA is an authentication method (akin to asking for a password, albeit a special one). It's not a digital signature.
So I usually start with "what if the back rotors go a little faster?": it tips it forward, beginning a series of operations leading forward motion (like turning the wheel of a car away from center -- what happens if you keep it turned?)
I've also found that words matter: "attitude", "differential", "pitch", "velocity", "acceleration", etc. -- may be hard to access, and simpler terms like "faster", "slower", "speed", "tipped" are usually more broadly understood.
"Can anyone name a type of engineer?" and for each shoutout, talk a little about what each does. The guessing keeps everyone's attention.
Types: software/computer, electrical, mechanical, chemical, environmental, civil, nuclear, aeronautical, etc.
Bonus: bring 2-3 props (your drones would be GREAT) that you can hold up for the relevant flavor.
For drones, I like to ask: "how do you think the drone moves forward?" [over power the rear two, underpower the front] Same Q for other directions.
"how do you think it rotates?"
"why don't all props spin in the same direction?"
Bring one of these to fly in the room: https://www.amazon.com/Cheerson-CX-10-Diameter-2-4GHz-Quadco...
Talk about how there's actually a computer on board.
If you have access to a TV, bring a short ~2min drone POV video to show.
Tell bad jokes ("Civil engineers are very nice to each other", etc.) Google online for some.
It's all about energy and making it interactive.
Pareto distributions are common in two-sided markets with geographic factors. BUT, when there's a reasonable possibility of extra regulation imposed at the regional level... that's a big risk.
We are all solving for different things.
Some want a shot at changing the world, some want to get rich, some want a job that (mostly) "doesn't come home with them", and many want to trade off between stress+anxiety and income+convenience. Others want to be always learning. Some have families and non-work commitments, others don't.
For example:
>>In my experience, it’s harder to have any visible growth in a
>>startup because well first, what do you grow to? There is no
>>clear hierarchy, no promotion paths.
In startups, there's often a better chance to learn general skills outside of (say) just development (dealing with users, writing content, devops, QA, teamwork, etc.) for those that have initiative.
Larger companies can be more structured but may have a chance for a more defined impact, with an existing customer base and market.
Which makes more/less sense? Depends entirely on what YOU care about.
It can be incredibly time-consuming, leads to a bunch of stress and uncertainty, and you can’t really bill for it.
In 2019, it's very hard to do anything "interesting" with BASIC when kids are surrounded by apps and games with hundreds or thousands of person-hours of development. "Hunt the Wumpus" just doesn't cut it anymore.
As others have written, Python is much more interesting because of the extensive library ecosystem. Integrating the physical world, graphics, controllers, sound, audio, network, video, etc. offers a lot more engagement than "Hello World".
Javascript is also interesting because it runs everywhere, and HTML/JS skills will offer good summer job money. :)
On the positive, it's generally good to raise money when you don't need it. You're more likely to get the most favorable terms. (Or, stated diffrently, it's hardest to raise money when you need it most, with a zero cash date looming).
On the negative, it creates a high bar to clear for the next round of funding. If you don't raise enough, and need more funding later, doing a down round is extremely painful. Worse, if you have anti-dilution and other provisions, the cap table gets blown up. Lots of agita.
Public companies: different circumstances (market cap is a currency for acquisitions).
Finally, I'm respectful of audacious and aggressive entrepreneurs, but I've got a very sensitive hubris detector. A billion is an arbitrary ego number, and "We don’t entertain offers.... . . and it’s literally not worth the time of day, talking" is crazy. Tell that to Groupon shareholders that held stock during Google's (reported) ~$6b offer in 2010. Ouch. Pride goeth before the fall.
It’s entirely self service.
Governments are sometimes just like people -- they get addicted to revenue streams and have a hard time adapting when those revenues change or go away.
Since it’s a government service, it’s entirely free and will likely stay that way. I have noticed occasional data quality issues (e.g. missing items), and I recommend a catching/proxy layer if you are planning heavy usage.
That’s one model that’s clearly working at some scale, and it could be easy expanded to “subscriber only videos“
A functional style is not the only way to write programs, but it’s often much easier to reason about behavior when you’re not chasing side effects.
And OO implementations often have a LOT of side effects.
While I do feel that folks grossly underestimate the skills that the best carpenters acquire over time, the software ecosystem generally is unique is that it's vastly large and complex. By any measure (switch points, surface area, person-years, etc.), the software we've all built in the past 70-ish years is the most complex thing ever implemented by human-kind.
Long ago it became far too large for any one person to "master" to the same degree an excellent carpenter might master his or her skills.
And the market agrees with the point of view, and values the economic output of the "best" software developers (at scale, not just one data point) at a far greater multiple over the "ok" ones, relative to other skills and trades.
One may not like this analysis (and my associated statement), but it is the state of the world.
TL;DR: Proposed machine costs $3k plus ~$100-200?/month for electricity, comparable AWS is currently $3/hr.
My conclusion: If you're going to do more than 1-2,000hrs of GPU computing in the next few years, start thinking about building a machine.
And the irony of this entire writeup is that number theory is now beyond "useful", as the underpinning for cryptography and security in our modern systems.
RT5350, 360 MHz, MIPS 24K, with 32MB of RAM.
CPU datasheet: https://www.mouser.com/ds/2/813/RT5350-1022839.pdf
I said: “Facebook‘s biggest competitor is the US government.”
I feel that’s even more true now, and the same comment applies to Google and Amazon. It’s not a coincidence these firms have massive lobbying efforts.
The cryo guys had something faster running in the back lab, but I don’t think they got 2x.
Indeed. 300mhz, dual-issue...in ~1991! Of course, that was before low-power was "cool" and the first chips were very nice space heaters.
Source: worked on the Alpha team before it was "Alpha" ("EVAX").
From the docs:
"Crystal is a programming language with the following goals:
Have a syntax similar to Ruby (but compatibility with it is not a goal).
Be statically type-checked, but without having to specify the type of variables or method arguments.
Be able to call C code by writing bindings to it in Crystal.
Have compile-time evaluation and generation of code, to avoid boilerplate code.
Compile to efficient native code."
And it's very difficult for even the experts to balance the positives against the negatives: facial recognition could be used to find a serial killer ... or a missing child.
As one example, RMS has long advocated against the "ethics" of certain technologies and companies, while Emacs and GCC have greatly accelerated the adoption of technology, generally. And those tools have almost certainly been used directly or indirectly to build some very, very bad things.
See: https://payne.org/blog/dividing-founder-equity-in-the-very-b...
TL;DR: (a) figure out ratios of ownership in the beginning (vs absolute percentages), and (b) use simple "buckets" of contribution to figure out the ratios should be.
For example, the two founders that have been working on the idea for 6-12 months may end up with more equity than the next "founder". Or a part-time (e.g. academic) co-founder may get a little less.