Nuclear power is expensive. Desalination water is expensive because of the electricity cost. Are you proposing to raise the cost even further?

Let's just take this at face value, and let's falsely pretend that the water shortages aren't also causing cutbacks by residential consumers. Then that's 40% of the state's water to provide a fraction of 2% of the GDP, in drought conditions.

Even in your scenario, do you think this is appropriate?

I wonder if they could help increase the economic picture via value-added product recovery from the discharge brine. The oceans have an interesting mix of dissolved minerals and there's already interest in recovery of a number of them; perhaps the concentration of the discharge brine could help improve their economics a bit.

(of course, what I find a more interesting possibility for recovery is mining the pacific garbage patch for minerals that have over the course of years soaked up into the plastic from the seawater)

Agriculture is the big culprit, taking 80% of the state's water (and in return ag and mining together only make up 2% of the economy). Its a totally unsustainable situation that has to be remedied sooner or later.

That said, I do have hope for the future of desalination. Not with current techs (as with the one in the article, they're energy hungry and expensive), but potentially with new techs that don't rely on electricity as their power source. One I find interesting is this one. Basically, it relies on evaporation, which isn't unique... but *not* by capturing the evaporated water. It's just concentrated salt solution that's desired, which means that you don't need some sort of elaborate vapor capture system and sealed tanks, just simply any sort of open area that can hold water - even an endoherric basin or jettied-off chunk of ocean. Far, far cheaper.

Concentrated brine is turned into freshwater via ion bridges: it's connected to two tanks of normal seawater, one by a positive ion bridge and the other by a negative ion bridge. The brine greatly wants to dilute into the normal seawater, but it can't because the ions would be imbalanced in the two side tanks. So these two side tanks are connected to a third tank of seawater with the opposite ion bridges, so that salt can dilute from the brine into the two seawater tanks, but only if they also "suck" the opposite ion out of the final seawater tank. Since the brine concentrated brine wants to dilute so much, the action is energetically favorable and continues until there's no salt left in the third tank - aka, it's freshwater. (An actual implementation would be a continuous process, not fixed tanks, of course)

Apart from basic pumping needs, there's no electricity needed. The energy source is just "sun falling on any water chunk of seawater that's not free to circulate with the open ocean". You might even be able to have it filled automatically in some places via the tides or waves breaking over a jetty without having to pump new seawater in, leaving the only pumping needs for distribution.

Of course, the main tech limitation right now is making the salt bridges have high enough throughput and reliability to justify the capital costs.

ICANN allowed it and now they're back pedaling since these brand and trademark holders pretty much have to purchase their brands in every new gTLD anyway, so the ICANN fee of 18-25 cents per domain really adds up.

Many registry operators have them, they are called "blocks" where you put a block on your TM'd string like "slashdot". For example, the Donuts registry which has over 200 new gTLDs allows you to buy a "block" which applies to all their TLDs for a fairly reasonable fee (a few hundred dollars).

.sucks does have blocking... but it kinda sucks (-:

I'm not sure where you got your numbers from, there are only 919 root-delegated Top Level Domains. There are a few hundred more pending new gTLD application with ICANN so the total for the next few years won't exceed 1200. (There are plans for a second round of new gTLD applications. The first round cost each applicant $185,000 USD.)

Definitions:
TLD = Top Level Domain
gTLD = Generic Top Level Domain (.com, .net, .org, .info, .biz)
new gTLD = New Generic Top Level Domain recently allowed by ICANN (.club, .bike, .software, .guru, .ninja, .computer, .sucks, .wtf, .porn, .xn--io0a7i, .google, .canon etc etc)
sTLD = Sponsored Top Level Domain aka "restricted TLD" (.aero, .pro, .tel, .museum, .travel, .edu, .coop etc)
ccTLD = Country Code Top Level Domain (.uk, .me, .io, etc)
Extension = a sub-domain you can register under (.co.uk, .de.com, 0.bg, .com.au etc)

Sponsored TLDs are restricted. For instance, you need a "UIN" delegated by the "Travel Industry" for a .travel domain, only legit museums can get a .museum domain, and only licensed professionals can get a .pro domain, which is why you don't see many of them (and never get spam from them either).

All legacy gTLDs are unrestricted. For awhile, .info domains were sold super cheap ( $5) so scammers bought them up.

Most new gTLDs are unrestricted, while some are restricted like .berlin and .nyc (need to be local to the city) and .bank (need to be a real financial institution and get audited every 2 years and sign your domain with DNSSEC, etc).

ccTLDs can do whatever they want and are not governed by ICANN.

For now, you can "blacklist" new gTLDs without much consequence, because people and businesses are only starting to use them. Keep in mind scammers/spammers/annoying-people register CHEAP domains, so you might want to blacklist .xyz (cheap) but not .bank (expensive). But in the future, legitimate activities under new gTLDs will occur so you might want to allow them over time.

But really, why block at the TLD level and not based on content and RFC compliance?

You can get a $15 .sucks domains -- BUT it must be hosted on the registry's website, which provides a "moderated forum" for expressing speech about something you think sucks.

The $2500 for trademark holders is extreme relative to other new gTLDs. Many charge a few hundred dollars for "trademark enabled sunrise registrations" (where you must have a registered trademark with the ICANN approved Trademark Clearinghouse (TMCH) which costs a few hundred dollars a year to maintain).

Your first example sounds weird, and the second is unambiguously not valid English. Neither express exactly the same sense of action as the example. And the sentence I gave was just one example out of countless. Word genders are just word groupings which help reduce sentence ambiguity. You can detach them from the concept of a human gender; you could call them wizzlewozzles, it makes no difference. They just reuse the pronouns from actual gendered beings, which saves some linguistic space.

Also, the concept that one has to learn a gender for each word... first off, in most languages with genders, there's patterns for recognizing what words are what gender. They're not always universal, of course (in an ideal language they would be simple and universal), but they're important. Secondly, there's all sorts of data a person has to learn about each word to be able to speak properly as it is. Even if you have to learn a gender for each word, patterns aside, it's just a tiny piece in a much larger puzzle - that puzzle being reliably and concisely transmitting information across a noisy connection.
 

With three genders in Icelandic, the odds of a collision between an arbitrary two is only one in three. As you might say, "good enough for government work". Languages are never, and can never be "perfect", there's always tradeoffs on how you allocate space. But I have to say, your example of how to do it in English was a pretty bad tradeoff (having to repeat "the backhoe")

Corr: stirling, not sterling ;)

Unfortunately, from the air you can't actually do a whole lot more than from low orbit.

The whole point of "from the air" is that you can land. And take off and land and take off again and again. It's not exactly tricky to raise or lower altitude. ;) And on Titan, low gravity plus a dense atmosphere yields very low terminal velocities, i.e. cushy landings without much effort. But even if you're not talking about landing, you absolutely can do a lot more by virtue of being able to get so much closer to the surface. Much of science is about resolution.

Meanwhile, Titan is MUCH colder Mars, at around 94K average surface temperature, whereas Mars' average is about 218K . Building equipment to operate reliably in those temperatures starts to get difficult.

)

The issue isn't average temperatures. The issue is temperature ranges. Having to tolerate a single very cold temperature is far easier then having to tolerate temperatures that range from very cold to very hot. The biggest challenge for Huygens wasn't tolerating the cold of Titan, it was tolerating the cold of Titan *and* the solar heat from flying near Venus.

The average day-night temperature on Mars is a 100K difference. Seasons add another 45 on top of that. This repeated thermal cycling incredibly stressful on all of a rover's systems. It also mandates a much more complex thermal management system. For example, Curiosity uses a system of computer controlled pumps to pump a cooling fluid from its RTG either to the probe internals or to the radiators based on the ambient temperature. If it wasn't an actively managed heat flow it could overheat the probe in the summer or freeze the probe in the winter.

Remember, it was ultimately winter that killed Spirit. Had the temperature been steady, and Spirit been designed to handle that steady temperature, this would not have happened. When you have a very specific temperature envelope, it's just a situation of "insulate and add radiothermal heaters or heat output from the RTG wherever needed". It's a lot more complicated when it fluctuates.

And contrary to popular belief, it doesn't get harder to heat as the temperature gets colder, it actually gets easier - that is to say, the amount of energy needed to heat a craft to 300K in a 100K environment isn't dramatically higher than the amount of energy needed to heat it to 300K in a 200K environment, far less than double (the exact details depend on the situation). Radiative heat loss is proportional to the temperature to the fourth power, so it drops off to nearly nothing at low temperatures (we'll just ignore for now that on Titan you're also not exchanging heat with the vacuum of space, rather the upper atmosphere). Convection on Titan is also greatly slowed. That's why the calculations show that you can heat a whole sizeable hot "air" blimp on Titan with just an RTG (that, and how much lift you get for how little effort!).

And as you point out, nuclear is really the only viable power source, which means you have to figure out how to dissipate waste heat without damaging the surrounding environment beyond your ability to learn anything.

Yet people use RTGs aplenty - even on Mars where it's possible to use solar. Almost half of successful actively-powered Mars surface probes have been RTG powered - the two Viking landers and Curiosity vs. Pathfinder, Spirit, Opportunity, and Phoenix. RTGs are more reliable, simpler, and provide useful heat. But they're also expensive, which is the main reason for the inner / outer solar system solar / RTG dichotomy that Mars straddles. A typical high power RTG may cost on the order of $100M. Not a big deal if it's a Flagship mission, maybe a big deal if it's a New Frontiers mission, and definitely a big deal if it's a Discovery mission. It'd be nice if a Sterling version could be finally developed, that'd be another big facilitator for outer planet missions by bringing down the fuel cost per unit power.

Avoiding doing damage by heating the environment has never seemed to be challenging. If it was a big problem, you'd be seeing Mars rovers radiating their heat out on boom arms (a simple solution to the problem, albeit with some potential mass penalty). But that's never been necessary thusfar. On a Titan flier, one would expect radiators to be on one end (say, the tail of a fixed or tilt wing aircraft) and sampling conducted on the opposite end.

For fun. Why not humor the submitter?

To the submitter: Okay, I'd start with saying, "don't reinvent the wheel more than necessary". So for example, consider IPA as the writing system. Or if you want to invent a writing system to be optimized by a given set of rules, at least consider using the IPA forms as your basis.

Consider who your target learners are. Is it the whole world? Any particular weighting that you want to apply with certain native tongues? Check and see what phonemes and linguistic rules are common in the languages by whatever weighting you want to apply.

When doing your weighting to decide what phonemes to use, don't only consider "whether the language has it", but also "how easy is ot for people to learn who don't know how to do it. For example, among the sounds in Icelandic that aren't in English there's the "ll" lateral plosive and the alveolar trill "r". The "ll" is nothing like anything found in English, yet given a simple description most English speakers can pronounce it perfectly. On the other hand, some people struggle for years and never manage to learn a trilled "r".

That is, all to say, an ideal language takes research not just on what phonemes people use, but what phonemes are easy to learn.

Then there's one of the biggest issues, which is intelligibility. You want the most diverse array of phonemes possible without being likely for the listener to confuse two similar ones together. Again, research would pay off big here.

The exact same rule applies to vocabulary / grammar, and this is unfortunately one thing that constructed languages usually suffer from relative to evolved ones. If English had the word "dog" like it is now, but the word for cat was "dawg" with only a slightly different pronunciation, these two common everyday words would lead to a lot of confusion. This normally gets steadily selected out either with pronunciation shifts or the adoption of alternative words.

If you really want to get into it, you could write an evolutionary algorithm to optimize your vocabulary and/or grammar to maximize the auditory difference between different common words and word phrases. The goal is to keep that signal to noise ratio up to maximize understandability. :)

One I'd recommend is something that Icelandic does: having a simple, universal stress rule. That is, the first syllable of every word, and the first part of every compound with at least one syllable between them, is stressed. And when I say stressed I mean literally double the length of the others. What this does is make it so that even a beginner can tell exactly where one word or part of word ends and the next begins.

A couple things that English speakers often attack about other languages you should think about instead of just readily dismissing them:

1. Genders. It seems archaic, right? But there are practical reasons. For example, consider the sentence:

"I used a backhoe to drag a box but it was ruined in the process"

Which is ruined, the backhoe or the box? In Icelandic it's obvious because a backhoe is feminine but a box is masculine. Sorting words into differing groups adds some clarity to sentences. It comes at the cost of increasing the amount of knowledge needed for each word (this is usually done by breaking words into patterns, such as "if it ends with these letters, it's in this group"). You could, for example, have such a grouping (calling it something other than gender), but have the rules for determining whether a thing is in a particular group be really obvious. Taking a direct from English example, if we wanted many groups, one for each last phoneme in the word, the above could become:

"I used a backhoe to drag a box but itoe was ruined in the process."

Now it's obvious to a "native" speaker of our constructed language that the particular word for "it" refers to the backhoe.

The other thing English speakers often complain about is declensions. But once again, they're another example of giving additional information to help the brain understand the lossy data stream coming from our ears, at the cost of using up part of the linguistic space. For example, in Icelandic, "skila" ("to return", pronounced "skihl-ah") takes the dative while "skilja" ("to understand", pronounced "skihl-yah") takes the accusative. If you're in a noisy environment, your brain may miss the auditory cue as to whether the word said was "skila" or "skilja". But when you hear the following word in the accusative or the dative, it automatically cues the brain in, "Oh, that's what they said!"

Again, if you do any sort of declension, just try to be regular. Declensions are not inherently bad (in many ways they're often just postpositions glued to a word) - they're just additional information in the data stream to combat the poor signal to noise ratio that we have to deal with.

Be sure to plan in advance for people to slur any words or particles together where they occur commonly together. That's going to happen in any language. Try to design the language such that it can happen gracefully. :)

Try to have as many regular "forming patterns" as possible. For example, in English, one can make a word meaning "having the property of X" of the form "Xish" or "Xlike". There's a huge number of common patterns with which one could form words, and the more of them that have regular rules, the better. "One who does X to Y". "One who Xes." "An X that can Y". "The act of Xing". "The act of X-Ying". And on and on and on. It might pay to just spend a long time reading the dictionary in random order and looking at the simplest possible definition for each word and seeing where common patterns for formation might be possible. The more patterns there are, the easier it'll be for a beginner to figure out what an unknown word will be even if they've never heard it before.

Lastly, I recommend reading about as many different languages as you can, because so many languages have so many dramatically different features. And I guarantee you, you'll run into some things that will make you think, "Oh, that's neat! I should totally do that in my language!" Things that you would never occur to you that a language might do. The way different concepts are broken up can dramatically vary between languages.

You probably came here thinking "how can I make the words familiar to people?" But really that's the last of your concerns. Your main concern is to keep the signal to noise ratio up, to avoid collisions and provide the brain data for error correction. That's how a language becomes workable. When you've got the choice, you should correspond with your learners' languages as much as possible, but don't sacrifice clarity to do so!

Again, Captain Strawman strikes. I am mortally wounded!