vqc

@chrisco23 can I trouble you to send me an email at vince@chessrapid.com?

I've fallen into the chess craze.

I put together www.chessrapid.com for my own use. It's a version of Puzzle Rush / Puzzle Storm that has the added benefit of: 1) filtering by tactical themes and 2) seeing what themes you are frequently slow on / incorrect on. This way I can narrow in on the tactical themes that I need to improve pattern recognition on (e.g. I can take a break from Mate in 1 and Mate in 2; I really need to work on middle game).

@chrisgd what's the best way to reach you / can you ping me at the email in my profile?

I imagine something like a local text file / draft email (because I always have my email open)?

The problem I run into is that I'm either running from thing to thing so there doesn't seem to be much time to write things down as they happen OR I run out of steam at the end of the day and I don't prioritize "write things down".

(cofounder here)

We've explored a bit of the ASL space. The motion based ones are not something we've dug into. But some of the "static-but-location" based ones are things we've tried.

"Mom" is easier than "grandmother". We've tested out a version of "mute". "Tile" would be quite a bit harder (I think).

(cofounder here)

This is a really great point. We do have some early thoughts around the idea of the interface being responsive to "distance from user" as opposed to "screen size". The chat interface, for instance, will change in size depending on how far away from the device the user is.

This doesn't change the particular problem you noted, but the general issue is an interesting space to dig into.

(cofounder here)

Ten feet seemed too ambitious and five was too many letters =). We didn't think about the six feet under connection, but now we'll have to figure out what to do with that. Probably lean into it.

There was another comment that was less convinced about controlling with the phone. Is there something about that modality / use case that you are particularly moved by? When you imagine "controlling" by phone what spaces would you want to dig into more?

Thanks!

vince@wakasaba.com

(cofounder here)

> - using my hand as a mouse in space-constrained situations where I can't have a physical mouse

Are you thinking that you would consider using something like this to interact with your entire computer generally (i.e. instead of solely in the context of a video streaming like shown in the demo)?

> - dance class, so the instructor can seek through music, especially for rewinding to the same starting spot over and over again. (This is from the days of in-person classes; it drove me nuts that the instructor had to keep walking back to his phone and then getting it a few beats off)

This is really validating feedback and affirms a lot of our experience in these types of classes (especially remote). Requiring the instructor to walk towards the device to basically do anything really messes with the flow of the class. And requiring the learners to walk towards the device to do anything basically reduces the social aspects of the class to zero.

We actually do have a built out music control system that we took out for the purposes of this demo.

We also have voice controls that can be used to control the music playing (among other things).

> - spellcasting game

This prompted a considerable amount of discussion during our most recent sync.

If you're available / interested, can you reach out to vince@wakasaba.com, we'd love to chat some more.

Thanks!

(cofounder here)

We're not too familiar with toastmasters (other than hearing that it's a great way to learn public speaking). Would you find this useful in the context of in-person or remote meetings or both?

And what would it be used for? Moving slides? Displaying / hiding the agenda? Sending lots of clap emojis?

Thanks!

(cofounder here)

Yes! The sudden increase of webcam usage is one of the things that prompted us to think about opportunities around this type of interaction. Sometimes I wish I didn't have to talk to my Nest Hub in order to make it do something.

(cofounder here)

@saulrh,

To clarify, are you describing something to the effect of: [whatever WakaSaba is doing] => Browser Extension => Webapp version of $EnterpriseVideoconferencingSuite?

This is not something we have explored. If you had to hazard a guess, what $EnterpriseVideoconferencingSuite would be most amenable to this sort of work?

Thanks!

vince@wakasaba.com

Fair enough. Palo Alto is a nice area, expensive market.

What kind of house is a reasonable house for $2.5M?

I've never heard of anyone who hit their billable minimums (e.g. 1900, 2k, etc) not get bumped to the next year. Although I guess there's a bit of a dance that happens years 5/7-9 for people who think they have a shot at partner.

Are you sure this is true?

It seems -to-edit many more than one or two of this hundred https://www.vault.com/best-companies-to-work-for/law/top-100... are still paying lockstep.

Cravath still lists lockstep raises for years 1-7. Skadden years 1-8. Latham 1-9. Davis Polk 1-6. Etc. Are the firms reporting one thing but actually doing something else?

This is not the exact post I was thinking of, but the one I'm thinking of is definitely an extension of this one. Having this post by jeff will definitely help me on my search. Thanks!

Can you give an example of how AI might help judges/lawyers/etc discern facts and avoid the influence of rhetoric?

Isn't it a bigger issue that even if people agree on the facts, the law can be mushy with concepts like "reasonable" such that two people can reasonably disagree on what is reasonable given the same facts?

This might be one

"Systematic review of safety checklists for use by medical care teams in acute hospital settings - limited evidence of effectiveness" (2011) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3176176/

Has there been any updates in the past five years?

could also have switched the verbs, i.e. "but will continue to advise the organization and donate"

The application now comes in two parts: one for the "business" and one for each founder. I believe the "hack" question is now in the founder application.

The Bay Area is extremely bio rich. It is on par with Boston and SD w/r/t bio jobs, lab space, NIH funding, VC funding, etc.

Consolidation is also a big issue. Regional specialty hospitals are buying up local GPs. Companies like Banfield are also buying up specialty hospitals and GPs. This reduces the number of software buyers and obviates the need to "transfer files" because all the files are already within one system.

We originally believed all the reasons sudosteph points out as being a good reason to enter the vet tech space. It didn't quite work out for us.

Source: Previous startup helped vets/owners transfer records electronically.

The founders did leave a mechanism for amending the constitution. It is very difficult to use.

Do you think they underestimated the difficult of amending the Constitution using the process they suggested? Or maybe they didn't foresee having SCOTUS be the ultimate arbiters of the Constitution?

I think the problem might be that plain words are still really spongy? A simple statement like "no vehicles allowed on public grass" isn't actually all that simple.

What happens when the words of the constitution apply to 95% of a given situation? 85%? 10%? Is it better to amend the constitution every time there's an edge case?

A lot of time you can just put a catchall textarea to cover all the rest of your input edge cases. For the law, that textarea is where all the action happens.

But only kind of. Even if you could get a strain of yeast to produce a specific antibody 100% of the time, how reliably could you get the yeast to grow? If you can get it to grow, how reliably can you extract a enough of the antibody to form just one dose? If you can extract, can you get it into a formulation that is pharmaceutically viable? Its not just about synthesis.

There's so much tooling and expertise that goes into front end, back end, deployment, db, QA, etc ,etc, etc just to get one app going. Why would would drug deployment be any different?

It depends on the scope of the patent. But to some extent, kind of. See, e.g., the patent for an adalimumab (Humira) formulation https://patents.google.com/patent/US9085619B2/en. The patent discloses materials and methods for creating a formulation of the drug.

The real issue is probably that there are so many moving parts, especially when you get to the more complex drugs (e.g. antibodies). You would have to go through a million patents to find all the information required to create the actual molecule and also get it in a state to put into your body.

And even with all the information, it is unlikely anyone is sufficiently resourced (whether it's money, tools, or know-how) to actually make something viable.

Really? I didn't know federal judges can request cases. Is that specifically a N.D. Cal. thing? I always thought cases were randomly assigned to a judge in the relevant district.

I mean, he went to two "Top 50" schools studying math. Do you mean he didn't go to an Ivy League or a school traditionally known as a top CS school?

I found

  Introduction to Machine Learning with Python - A Guide for Data Scientists (Andreas C. Müller, Sarah Guido, 2016)

to be a good introduction to using and tuning these algos.