milspec

That's great for marginal costs. What about fixed costs?

The lobby needs to be mopped. The parking lot needs light. The lawn probably should be kept from turning into a forest. The heat should run in winter, and the cooing in summer.

Which hospital users get stuck paying for those things? What is your fair share?

12 mL worth of battery ought to be able to handle most of those methods. That's about break-even, but better because you refill the same way you refill the computer and propulsion. Charge the battery and you're ready for more action, mess-free. If you can do the job with less than 12 mL of battery, it's a definite win.

That does seem kind of neat, but it still doesn't make sense to me. You have to carry a poison supply that will run out. You need 12 mL per starfish.

Is that really more effective than just injecting lots of seawater? You could pump in seawater by the kL, not mL, and you'll never run out. The starfish will surely die after all bodily fluids have been washed out with seawater.

The other idea that comes to mind is a needle with an electrified tip. Depending on voltage and current, this kills in various different ways. You can heat the animal, zap the nerves, or (via electrolysis of salt) fill the animal with either chlorine or sodium hydroxide.

The really sad thing is that we build on our best farmland. If you want to grow apricots, the best place is San Jose. That whole area in fact is mighty good for farming.

People need to eat! Californians ought to be living in the foothills of the Sierra mountains. This puts them close to water and saves the more farmable land for farming.

Good farmland is flat, is not full of annoying boulders, and has nice weather. We thus pour slab foundations for giant 1-story McMansions and dig holes for giant swimming pools. Doing that in the foothills would be a short-term expense, so instead we'll destroy our best farmland forever.

You missed the word "repeated". Also, some participants in drunken fights act in self-defense or as part of a fighting sport, and so would not be criminals.

So it's actually like this:

You reach 25. You assault somebody. You serve time in prison. (now a questionable case) We decide to give you a second chance. You feel no need to follow the law, and you enjoy assaulting people, so you do it again. OK, at this point we know what kind of person you are. You didn't just make a bad decision one day. Simply letting you out into society is pretty much the same as assaulting a random person, because that is clearly what you are going to do.

If the offender has only ever committed one crime, and they did so while a teen, then OK. Give them another chance, provided they have a workable budget that includes a non-criminal job.

If the offender repeated the crime while being older than 25, then no. There is no realistic hope that the person will ever be non-harmful. Never let them free.

(cases between those two extremes are questionable)

Standard algorithms are standard for very good reasons. They are 100% reliable and at least reasonably fast, if not the fastest.

Failure to learn the standard algorithms impedes progress. You are unlikely to do well in the next course if you are slow and can't handle numerous cases.

Learning alternate algorithms is great... once you have the standard algorithms down solid and are ready to prove equivalence.

Japan does this right: nobody tips. Customers pay the restaurant, and the restaurant pays the waiters.

Tipping is one small step away from paying bribes. Don't do it. Don't encourage it. Don't accept it.

That's a tough one. Maybe you could help work on a JIT for Javascript or Java. This would get you into working on the low-level stuff while still making use of your existing knowledge. Another option is to pretend you are back in school. Find a course with syllabus and homework assignments posted online, buy the book that goes with the course, and do all the assignments. The aggressive approach is to just sit down and write your own boot loader or maybe jump into a capture the flag (CTF) problem with IDA Pro. In case you want to learn from a simple OS, look into xv6 from MIT. Add a few features, possibly as suggested for various courses that use the OS for teaching. You could add a debugging interface. Another idea is to write a disassembler.

It's horribly insecure to keep the company's source code on a computer with a connection to the Internet. For example, this is how the F-35 secrets wound up in China.

We wanted our own site because the parent company's branding was extra-uncool and they wouldn't change it for us, but they wouldn't let us use their name without controlling the branding. This was the compromise. Aside from a few issues like that, we are pretty distinct from the rest of the company. We have our own culture (unusually lively for a defense contractor, almost start-up-like) and we like to show it.

The locations are Florida, Texas, Virginia, and Maryland. Note that the Florida location would let you afford a big house in a decent neighborhood that is near both work and the beach. This is ONSITE. We hire INTERNS, experienced/old/PhD, and everything in between.

Most of our work is low-level. We deal in assembly language (assembler) for MIPS, ARM, x86/x64, PowerPC (ppc), MSP430, 8051, AVR32, and many many others. We write debuggers, disassemblers, emulators, hypervisors, static analysers (for both source and binary), and similar bug-finding tools.

You can run the OS of your choice. Overtime is fully paid and optional, so you can earn more if you wish or just enjoy your hobbies/family/sleep. Extreme flex-time lets you wake up late and/or run errands during the day.

http://advancedsecuritylabs.com/

We have a rockstar programmer.

No, really. He put out CDs and all. Technically he's a former rockstar I guess, but close enough. He rocks.

We still lack a ninja. We have been unable to find one who could do low-level programming and get a US (not Japanese!) security clearance. So far we've had to settle for Nerf dart skills, which really isn't up to ninja standards.

Those places are silly. We'd hire you if you can do low-level stuff. (background check too, but I guess you'd pass based on other posts)

* (If anyone is looking for a smart C++ dude, preferably someone using QT, lemme know, I'll hook you up.) *

Uh, maybe. Is he OK with low-level stuff too? We do use Qt, but we might ask him to write a disassembler or a kernel driver.

Oh my. I know somebody who needs that job. Probably the "Georgetown, KY" on his resume is the difficulty, but he's willing to move. He can do C#, Java, and C++. He even has a traditional 4-year computer science degree.

Pass along his email if you can: [email protected]

Depending on the town and your specialty, you might be surprised. Not every place outside of the typical hot spots is corporate-fief. Specialization means you can have a good job market without the crowds causing traffic jams and running up the cost of living.

For example, you might not expect much good in Melbourne, FL. You'd be wrong if you happen to like low-level code (assembly, reverse engineering, vulnerability research, embedded development, etc.) or EE stuff. There are both larger companies (Harris, Lockheed, Northrop Grumman, and the http://advancedsecuritylabs.com/ part of Raytheon) and lots of startup-ups.

Since a specialized area (like Melbourne, FL doing low-level stuff) doesn't have too many nerds outside that specialty, the cost of living stays low. You can get a cheap house within a mile of work and there isn't much traffic.

I think I recall getting you an interview. (at http://advancedsecuritylabs.com/ in Florida) Did we not give you an offer? Did you reject an offer?

We could easily put you in San Antonio or Austin. We might even be able to do Seattle or Cambridge (near Boston), though those are tiny offices.

If I remember right and we did reject you, try a different office.

This is likely in Melbourne/Indialantic Florida. It could also be in Texas, Virginia, or Maryland. Note that Florida combines low housing costs, lack of an income tax, super-short commutes, and very little traffic. It can be a shock compared to what you may be used to. This is ONSITE. An active Top Secret/SCI security clearance is required. Any overtime would be paid; we don't make you work for free. You get flex-time and an office with solid walls.

We need a system administrator.

The users mostly run Ubuntu, but some run Windows and other things. The users are primarily software developers. You should be familiar with virtualization (Open Stack, VMWare, VSphere, VCenter), scripting, LDAP, DNS, security, backups, and running cables.

https://rtnsi.theresumator.com/apply/u2SS6D/Systems-Administ...

This is Melbourne/Indialantic Florida, Austin and San Antonio Texas, Maryland, Virginia, Greer South Carolina, Fayetteville North Carolina, Huntsville Alabama, Cambridge (near Boston) Massachusetts, Seattle Washington, and more.

Note that a few of those locations combine low housing costs, lack of an income tax, super-short commutes, and very little traffic. It can be a shock compared to what you may be used to.

This is ONSITE. We hire at all levels, including INTERNS. We can only hire US citizens due to government requirements. Our background check includes a polygraph.

We do lots of reverse engineering, vulnerability research, custom in-house emulation (with JIT and/or hypervisor), constraint solvers, and static analysis of both source and binary code. We work with numerous OSes and processor types. You could write the tools (emulator, JIT, hypervisor, debugger, static analysis, etc.) or you could make use of them to find vulnerabilities. An understanding of compilers is often helpful.

You can run Linux. We don't expect overtime, but we'll pay you for it if you want to do it. Most locations have extreme flex-time and offices with solid walls.

If you happen to be at DEF-CON 23 in Las Vegas (August 6-9) you can catch us there.

Extra keywords: assembly assembler IDA Pro MIPS x86 x64 ppc PowerPC ARM ROP security VMX decompiler kernel driver embedded C99 buffer overflow opcode ASLR NX SMEP SMAP MPX 8051 MSP430 AVR32 MCU TOP SECRET TS/SCI SSBI classified clearance cleared formal verification sandbox fuzzer fuzzing crashes

http://advancedsecuritylabs.com/

A sundress is much closer to a suit than it is to a T-shirt and jeans.

Considering modesty though, a woman in a sundress is equivalent to a man wearing a tank top. We don't get males in the office wearing tank tops or anything else that would expose the collarbone, armpits, shoulders, or shoulderblades. These things are normally considered inappropriate. We also don't normally see males wearing snug form-fitting shirts. It's not considered appropriate. Feel free to wear that stuff while you weed your lawn at home.

So that is two strikes against the sundress. It's kind of formal, yet no more modest than a tank top.

The comment about being women being treated differently when they wear a dress shows misunderstanding. The male equivalent to that is not a T-shirt and jeans. It's the suit. I'd sure be treated differently if I wore a suit. I'd be trading one kind of signaling for another, giving up technical cred in favor of business cred.

Oh, one other problem. The referenced (R) and dirty (D) bits are only updated in leaf nodes. I strongly suspect that any performance you gain here will be more than consumed by OS software needing to scan all the leaf nodes for these bits. If you update non-leaf nodes, then the OS can use those to avoid checking many of the leaf nodes.

Thanks. The following concerns paging, not the base/limit system:

From a security and reliability perspective, I'm saddened to see that rwx got supported while --x did not get supported. That is backwards. Having to change permissions after code modification is not bad; this provides a convenient point for cache flushing and ASLR-enforced address changes. Preventing executable code from being misused as data is valuable.

I'm also saddened to see that user access implies supervisor access. This too is exactly backwards; nothing should be both user and supervisor accessible. Given that data access can be performed at a less-privileged level by setting MPRV=1, the ability of the supervisor to access user pages normally is especially strange.

Lack of distinct did-execute and did-read bits is mildly annoying. If a page is marked as being accessed and executable, one must assume that it is now in BOTH the instruction cache and the data cache.

I have mixed feelings about having page frame numbers shifted over by two bits. The win is Sv32 getting a reach of 16 GiB. I suppose this is worth the minor annoyance when debugging OS kernel code.

Other than that, I like it. It's certainly sane. The traditional page table is pretty good for the middle bits of the virtual address. I think it is less good for the upper bits due to ASLR, and I hate to see anything that encourages a failure to use all 64 bits of the virtual address space.

MMU?

I see one place seemingly using RISC-V with an MMU in the classic desktop PowerPC style (Linus Torvalds posted a great rant about the stupidity of that MMU) and another place that is seemingly using RISC-V with an MMU that is very much like x86 (the paging part, obviously no segmentation) but with distinct rwx.

Which is it? Did this not get specified? Constantly changing the MMU greatly hurt 32-bit SPARC and PowerPC.

FWIW, this is good: Bits 0..11 direct mapped, bits 12..29 are x86-style page table tree node indexes that are hardware-walked, and bits 30..63 are software-filled like MIPS. (a forest of trees) In the low bits of the bottom level you get: can read, did read, can write, did write, can execute, did execute, user/super (exclusive), type ram/framebuffer/mmio/pte (two bits), reserved, and validity. The "did foo" bits on PTE pages do get updated.

Underwater RF is no trouble in this case. You have the signal backwards. As long as there is no problem, regularly transmit an identity beacon. Loss of signal is the alert.

So, how'd that go over with the girl?

It's tough selling these on your own: http://weis2007.econinfosec.org/papers/29.pdf

The easy way, if you want to this sort of thing regularly, is to work for a place that discovers zero day exploits. You'll get a salary/wage and benefits. Here is a fun place to do that: http://advancedsecuritylabs.com/

Most anything you'd want to do with this is illegal due to FCC rules against encryption colliding with the need for computer security and now the recent push for https everywhere.

Yes, but did you ever mark mailing list email as spam? (actual spam even)

If so, maybe gmail thinks "this sender is spammy". That's the mailing list.