parametrek

Everything he wrote about that was wrong.

I was never accepted into the original Proposition. I did not get any money or advice. But my elevator pitch was good enough to convince me to go through with it anyway. And I'm hardly "floating in the software graveyard."

But I am glad you enjoy my site ^_^

> just serve them locally in a sufficient format and you'll slip through blocks.

In what universe do you live? Every little forum that self-hosts ads is already in the blocklists.

> The lead-acid grid lattice design, still the...errr...gold standard when it comes to the most amount of joules stored per buck, was invented in 1881.

Li-ion has caught up. If you have $400 to spend on batteries both li-ion [1] and comparable lead acid [2] (deep discharge, long cycle life) cost around 3 Wh/$.

[1] https://www.imrbatteries.com/samsung-29e-18650-2850mah-2-75a...

[2] https://www.powerstream.com/BBep.htm (EP100-12)

Those are still pretty spammy and often full of misinformation. Wirecutter's flashlight recommendations have always been terrible and so I have no faith in them at all.

Hi! To expand on #20 there is a lot of stuff going on but like you said most of it is SEO spam.

It is further hampered by planned obsolescence and the increasingly fast pace that new products are released at. It is very difficult and expensive to maintain actual experience with everything in a product space.

It can't be crowd sourced either because most people are overly attached to their purchases and can't give objective comparisons between something they don't own.

LiFePO4 is my favorite.

There are plenty of other lithium ion chemistries that use no cobalt. If push comes to shove we use them instead. I already use LiFePO4 everywhere I can.

Depends on the li-ion. There are chemistries out there right now that are good for 20,000 cycles like LTO. And a substantial amount of li-ion research is extending cycle life.

It doesn't matter how good hydrogen becomes. Battery tech will continue to improve faster than hydrogen tech. H2 will never catch up.

> HN readers are not the typical consumer. I guess the majority here will buy after making a comparative table

You are not kidding. That's how this was born: http://flashlights.parametrek.com

Channel your obsession! I'm the same way and it grew into the most comprehensive database of flashlights around: http://flashlights.parametrek.com

Technology hasn't kept up with that line. Here are 18650 cells on the market, sorted by Wh/kg:

http://batteries.parametrek.com/index.html?size=18650&wh_kg=...

The best are still only 276 Wh/kg, no where near the 350 Wh/kg the graph would predict.

> could you apply the same recipe for a comparison site to anything else with a numeric bang per buck quantifier

Absolutely. I've done this with flashlights, for example.

This is good work! I might reference your project in an article I've been writing about responsible affiliate sites.

Ooops, thanks!

While that is extremely light, it still weighs as much as 3x18650. Breakeven point would be after four days of perfect 100% utilization of solar power.

Is it practical even then? I'd gladly carry a few extra batteries so that I don't have to sit around camp and keep adjusting the panel every hour.

$100 isn't practical for most people either.

Your efficiency calculation is wrong. It isn't 25% of 300Wh, 300Wh is the output. The input is 400Wh, with 25% loss or 100Wh. Not 75Wh as you have. And you have a bunch of other mistakes too.

Along the length of the road, that is 21kWh. At 75% efficiency, that is 7kWh into the road surface per car. Roughly 2.4GWh of energy ($200k worth of electricity) per day lost as heat into the roadbed in total. Assuming 4 lanes and a 3.7m lane width, route 60 is 1.7 million square meters of asphalt and that is 1.4kWh per square meter. (8x more than your number.) Still comparable to solar irradiation though. Asphalt might get a little soft on a busy summer day when everyone is also running their AC but it wouldn't be a molten puddle. If the chargers are only embedded in a narrow strip down the middle of the lane, that section might get very soft.

So at 75% I guess it is practical. But 75% is going to be extremely hard. And the first place this should be installed is Route 1, because it would be a highway-sized snow-melter that also happens to power vehicles.

Now ask yourself why we don't have defrosted highways. Installing roadbed chargers is going to be even more difficult.

Not practical though. Every time someone has done the math for a fully scaled up system (assuming magic technology that doesn't exist) it always has the same answer: Unless the power transfer system is nearly 100% efficient, using it on a typical highway results in the the highway turning into a molten puddle in less than a day.

If you want to run the numbers yourself a Tesla uses 300 watt-hours per mile on the highway. California route 60 sees 337000 cars per day and is 70 miles long. And an amazing power transfer system might be 75% efficient.

They aren't worth the money or the weight. You are almost always better off bringing the equivalent in spare rechargeable batteries. I've written an article on it:

https://www.reddit.com/r/flashlight/wiki/solarcharging

Find a product that you like owning but have difficulty buying. Identify and fix those pain points.

No, parasitic is accurate for most affiliate sites. The majority are content mills who are better at SEO than the retailers. People google "best foobarbaz" when doing christmas shopping and end up at an affiliate content mill. If the affiliate content mill didn't exist, google would have taken them straight to amazon. This is highly evident by the numbers shared in affiliate forums - 90% of annual revenue often comes from November and December. (If you have useful content and are appreciated by the community, this does not happen.)

Then there are blogs who happen to use affiliate links. These guys are content creators who don't want to show ads. If you go this route you have to be careful to avoid becoming (or looking like) a content mill. But generally these are people doing good, tasteful work and only writing about products they've actually bought and put through the wringer.

Finally, in distant third, are the companies affiliating because they want to fix the online shopping experience but don't have the capital to compete directly. Camelcamelcamel is the best example of this. My own site is also a fair example.

There are many shades of gray in the affiliate spectrum. It mostly comes down to how many "best practices" (aka dark patterns, like making all images go to affiliate links instead of larger versions) you choose to use. If someone says their operation is parasitic, they are probably right.

I've mentioned this before [1] but affiliate sites work, particularly if you are already doing web dev.

[1] https://news.ycombinator.com/item?id=14473272

Thanks.

The pricing is the easiest part. As an affiliate you get access to an API that lets you query it. Of course it is heavily rate limited and sometimes fails outright. Accessing certain parts of the API comes with strings attached too. All of the non-amazon prices I refresh by hand every few months.

Web dev? Affiliate links.

Consider a consumer item you are passionate about. Walk through the typical shopping process, either as a newb or as an enthusiast. Identify all the pain points, every bit of friction. Consider everything you've listed as an embarrassment that you will personally make right. Build the smoothest, fastest, most respectful experience possible. Compile lots of information too - your goal is to become the best resource on the internet. Put the user first at all times. And when appropriate add affiliate links. (In my case only 27% of URLs are affiliated.)

Why is respecting the user the #1 priority? Because you do not want to look like an affiliate content mill! Go read the blogs that teach you how to affiliate^W build dark patterns and do the opposite of what they suggest.

Of course even if you are unquestionably the best resource on the internet, some communities will still tar and feather you for having affiliate links. I was very lucky to find a great and supportive community.

Amazon affiliate links. (Yeah, yeah, I need to diversify. Ain't easy making opportunities.) It works well if your site is designed to improve the shopping experience.

> guarantee

Of course that is not possible. At best you can grade someone's past behavior according to a set of criteria. Though in this case most of https://reddit.com/r/flashlight/ would probably say that I am trustworthy and impartial ^_^

> The other day, I wanted a super bright flashlight that ran on AAA batteries and was of decent quality and price. I had to pour over the millions of listed flashlights, most looking physically identical, with the exception of the brand name printed on the flashlight itself. Then I read about how these flashlights are all mass produced by the same manufacturer and re-branded

You need a trusted third party to collate all of the information about a product category and then filter out all the crap. I have conveniently already done this for flashlights:

http://flashlights.parametrek.com/index.html

(Yes, I run it and yes it uses affiliate links.)

Also, AAA cells are terrible for flashlights. A single AAA cells is okay for a compact penlight but many use 3xAAA cells and 95% of the time 3xAAA lights have the worst engineering of any flashlight.

Or you can design it to be light and fast from day 1. I had no trouble with the same level of traffic during a ShowHN on a $5 droplet.

'Always show checkboxes' was something I tried early on. It simply doesn't scale well. There are already more than 200 checkboxes and 10 double-ended sliders.

Each single checkbox presented is the most important feature for someone. It is interesting that the HN crowd is stressing high CRI but many feel the reduced efficiency is not worth the trade off.

Here are all the waterproof high CRI lights:

http://flashlights.parametrek.com/index.html?led_color=high%...

Though generally if you are serious about nice CRI you will end up buying a regular light and swapping out the LED.

It might be worthwhile for me to provide levels of detail. Like in your case, you don't care if something is IP-X7 or IP-X8, you just want waterproof. And no good manufacturer would say "waterproof" since the word doesn't mean anything. Though that sort of modal configuration screens (advanced mode, etc) are almost universally considered bad UI on HN, so who knows.

If you are looking for a flashlight I maintain the largest and most complete databases of flashlight specifications:

http://flashlights.parametrek.com/

Feedback appreciated if you think this has potential. UI is a little bit designed for the power user who knows what they like in a light. Still trying to make a gentle ramp-up mode for people with less of the jargon.

Sure. http://parametrek.com

I should probably re-shoot the demo video. It is seriously dated.