Hacker Newsnew | comments | show | ask | jobs | submit | aggie's commentslogin

A pragmatic approach would be to simply leave pharma alone, at least initially, as they actually need strong incentive for R&D.

-----


Or to socialize drug discovery using a competitive scheme akin to the XPrize with the reward being a free ticket through clinical trials and safety development (among other things).

-----


This is great advice. I would also recommend adding just a bit of additional synthesis to your summary if possible, connecting what the chapter/book talks about to other topics or ideas you may have. Creating strong and varied connections helps to build a robust schema in your mind.

I'm curious about your assertion that some degredation should occur before this activity. Can you elaborate on why? My thought is that it forces you to do a bit of reconstruction which aids in the goal of truly writing in your own words.

-----


So this was from the book "make it stick", recalling memory strengthens the network, which is why quizzing yourself or doing the simple sample problems at the end of the chapters are more effective then highlighting/rereading notes. The gap is mostly because you want what you've learned to settle into long-term memory where you can find it. Think of it as being a rat in a maze, you are on one side and the memory somewhere on the other, you want to work towards strengthening the path towards the memory, so it's easier to recall the next time around.

-----


Forgive my lack of a source, but I read a while back that hotel wifi is often so bad because they signed short-sighted contracts in the past that locked them in to these terrible data rates for many years. And now they pass that terrible deal on to their guests.

Edit: well I found the 'source' - turns out it was a Hacker News comment, so take it for what it's worth. https://news.ycombinator.com/item?id=8082519

-----


The other reason it's so bad is that they run hotels, they're not network engineers. Few people know how to set up wifi so it doesn't suck.

Most hotels don't ever have the IETF to stay.

https://news.ycombinator.com/item?id=3771876

https://www.networkworld.com/news/2012/032812-ietf-makeover-...

-----


See also ‘The “WiFi At Conferences” Problem’ by Joel Spolsky:

http://www.joelonsoftware.com/items/2009/10/08.html

[…] When you try to ask complicated questions and explain that your conference has a lot of techies, they say, yes, we understand, we have A-number-1 internet access, no problem very good. When you say, “Yeah, but have you configured your DHCP server so that it has more than the default 254 IP addresses available to hand out,” they have no idea what on earth you’re talking about, and of course it turns out that they had some vendor, a company you’ve never heard of, provide their internet access. And half the time, that vendor installed a DSL line from the local telco and hooked it up to a LinkSys WRT54g they got at Costco, then installed some kind of crappola welcome-screen software just to make it even worse, and then disappeared. […]

-----


Hotel managers aren't plumbers either, yet they seem to handle running water in their rooms without much of a issue. The problem isn't that they are hotel managers per se, the problem is that hotel managers tend to be older and so unable/ unwilling/unaware to adapt to new technology. I'd bet dollars to donuts that the quality of wifi has a strong correlation to hotel manager age for hotels that aren't part of chains.

Chain hotels, on the other hand, can fix this problem through the normal franchise methods: Determine best practices for the entire chain, and then provide standard instructions and help for each franchise.

-----


Don't oversell their plumbing ability. I stayed at a pretty expensive hotel (unfortunately built in the 1970's when the world became stupid with respect to construction) where the toilet exploded. I heard this gurgling noise, went to check it out... and there was old faithful!

The maintenance guy told me when the place was full, if people flushed simultaneously, "bad" things would happen.

Wifi is a similar thing. Best practice for Wifi is "it depends". You need to hire somebody with a clue and do surveys. Hotels have been traditionally build around sound and fireproofing, and the methods used to achieve both make Wifi difficult in some places.

At cookie cutter, modularly built hotels (ie. Hampton Inns, etc) Wifi is very uniform in quality.

-----


I' a bit disturbed by your open bigotry. You would never say "that hotel was bad value because JEWS". Why do you think it's acceptable to say "tha. Hotel wifi is bad value because OLD PEOPLE"?

-----


But as you point out, the IETF and NANOG have openly documented how to set up hotel wifi that doesn't suck. It doesn't take any technical experience whatever to hire a WiFi firm and instruct them to set up a system the same way that the IETF did it.

-----


This was brought up in a recent HN thread (that you just found).

The way it feels to me, forgive the profanity - it's assholes that got screwed by other assholes. Signing up for such "short-sighted contracts" is stupid, but offering those in the field of fast-changing technology (and not upgrading them when the technology changes to the point of the contract being ridiculous) feels... evil.

-----


I find it hard to believe that lobbying the FCC is cheaper than paying lawyers to engineer their way out of those contracts. With the crazy fees mentioned in the article for wifi access (seriously, $1000 per device?), it's not hard to see that their "protecting users" line of arguing is BS when there is such a clear profit motive.

Also, I totally don't get their strategy here, since it's not like there aren't other hotels which are nicer / cheaper / better-suited to conventions than Marriott. Really hope that they lose a lot of business for trying to pull such a stupid, petty trick to increase profits. In combination with their push to "remind you to tip their maids" so they can pay them less, I'm really excited to watch them burn.

-----


Pardon me if I'm skeptical that a multibillion dollar company the size of Hilton Hotels can't strong arm a local ISP into caving on a contract. The cost of a lawsuit alone would cause whoever said ISP is to shit their pants and give them whatever they want.

-----


While the veracity of the claim is untested, it doesn't seem so out there that many hotels that were at least partially forward-thinking may have not predicted the need for a better / faster network in the future...

-----


> although Jakob Nielsen's work has shown that you reach the point of quickly diminishing returns after testing with 5-10 users

This gets tossed around a lot, but is usually misinterpreted. What 5-10 users will get you is discovering the mere existence of most usability problems. What much larger sample sizes will get you is an idea of how many people are likely to experience that problem, which is very helpful when talking about trade-offs in meeting the needs of various types of users.

-----


A year is far longer than they would need just to achieve statistical significance. They've got over 160 million unique visitors per month [1]. Even showing the variants to only 1% of traffic you're working with over 50,000 visitors every day, enough to run large multivariate tests.

1 - http://www.statista.com/statistics/271450/monthly-unique-vis...

-----


Amazon probably also has vested interest in how a change affects LTV which is way harder to understand than whether there was a purchase regression in a short time window.

-----


Statistical significance is not the same as significance in business terms. I can't imagine anyone would use such an approach when deciding how much data to collect.

-----


>Statistical significance is not the same as significance in business terms

Depending on the level of significance, why not? If it's expensive or simply not possible to gather the full data set, sampling is absolutely a valid basis for business decisions. Why wouldn't it be?

-----


Maybe for some things, but not necessarily for everything amazon cares about, such as performance and reliability. For both of those amazon doesn't really care about averages, they care about the 3 9's experience. If there's a 0.5% chance that page load will take longer than some small number of milliseconds then that's not good enough for amazon and they'll go back to the drawing board. Factor in bugs in the new layout being fixed during the A/B test (necessitating resetting statistics) and it's easy to see how it could take a year to fully roll out a big change given amazon's cautiousness.

-----


Why not run a two-tailed test and double the alpha? If I'm understanding it correctly, you'll still make the same conclusion at either tail as a one-tailed test, but this way you have both directions covered. I could be missing something, just thinking out loud.

-----


The study doesn't show that writing on paper is necessarily more effective than typing notes on a laptop. The study shows that, when typing, students tends to transcribe verbatim, whereas when writing longhand they tend to paraphrase. This extra processing results in better understanding. This does not mean that you can't paraphrase while using a laptop, just that the students in this experimental setting tended to do that, presumably because they could type fast enough to transcribe verbatim and thus did so. So it's really just an issue of note taking style, not the medium.

-----


Good point. Though it's important to note that the reserchers identified an effect, and hypothesized about the mechanism.

It's possible writing improves understanding for some reason other than/in addition to paraphrasing.

-----


Weight loss is hardly as simple as learning to count, assuming we're talking about actually doing something and not just planning for it. You seem to assume very little capacity for learning basic nutritional facts and very high capacity for sticking to a diet (willpower). I would argue most people are in the opposite position.

-----


These are some great examples to get yourself thinking philosophically. If anyone reading would like to explore them a bit more, here are some wikipedia articles:

http://en.wikipedia.org/wiki/Qualia

http://en.wikipedia.org/wiki/Ship_of_Theseus

http://en.wikipedia.org/wiki/Trolley_problem

-----


The number refers to the capacity of working memory and has indeed been revised over time. It's a complicated issue, but 4 units of information is now considered a more accurate approximation. But if you're remembering a password, you're dealing with long-term memory, which doesn't have this limitation.

http://en.wikipedia.org/wiki/Working_memory#Capacity

-----


Long-term memory usually accepts most information from short-term memory not direct sensory input, so it is easier to memorize when you break down things in chunks first.

-----

More

Applications are open for YC Summer 2015

Guidelines | FAQ | Support | Lists | Bookmarklet | DMCA | Y Combinator | Apply | Contact

Search: