An important early impetus for research at AT&T was the immense labor component of contemporary switching technology. Hiring the number of qualified operators needed to scale out to a fully national phone system wasn't clearly possible, let alone sound business.
The scale of the challenge over time is described pretty well here:
AT&T thus had both a clear long-term goal, national phone service, and multiple long-term research challenges needed to make that goal possible (amplification)
and economically feasible (switching).
For folks interested in delving further into Labs history, this recent paper has a wealth of relevant citations, while also clarifying some important corporate cultural framing that might otherwise be confusing:
Yes. AIUI, as bicycle riders the Wright brothers understood the need for banking in turns rather than thinking of maintaining a constant roll orientation throughout a turn
The article mentions both 64-bit ARM and x86. Intel's ENDBR64 (part of CET) and ARMv8.5A BTI work largely the same. The instructions are NOP on older processors or if not enabled by the OS.
BTW. A similar extension is also about to be approved for RISC-V: "Zicfilp". It also repurposes an instruction that was previously a NOP.
So, ROP itself is a (frequently very effective) workaround for W^X memory protection features, where executable code itself should be unmodifiable at runtime. I wonder if there's a "next" CFI attack that calls existing functions that are actually defined as such, but in a weird order, or with weird arguments.
The ability to do that effectively would depend on lots of stuff, including the calling convention (as the stack or heap are easier to corrupt than registers).
Maybe someone is already writing an advanced ROP gadget-search that looks for "real functions that can be combined to make ROP gadgets by calling them in a bizarre way". (This is already much less likely to occur in a real program ... I think ...! But maybe the return-to-libc phenomenon is still a rich source of vulnerability?)
Interestingly, RISC-V's Zicfilp proposal includes an optional, 20-bit label operand so that call sites and targets can be paired more strictly.
I'm not sure how or if it addresses trampolines and similar thunks that interpose callee and caller, but which don't necessarily touch arguments. Such thunks are quite common in dynamically linked code as ELF and Mach-O do lazy loading by default--dynamic function symbols are initially small thunks that load the dependency and then forward the call, restoring registers and the stack without having to know the call signature.
There are fundamental design choices in Lua that can certainly put some people off, such as prototype-based inheritance or the union of arrays and maps.
It certainly wasn't casually hacked together, though. Ierusalimschy et. al. have thoroughly documented the language and its development rationale with admirable clarity.
In particular, the paper "C APIs in extension and extensible languages"[2] will be of special interest to prospective embeddable language developers. "Passing a Language through the Eye of a Needle: How the embeddability of Lua impacted its design" is also a good read, for those who can summon bits through the ACM hellm... er, portal.
The really troublesome securities of the Global Financial Crisis were complex derivatives of private-label, almost universally sub-prime, asset-backed-securities:
The standardized, so-called "agency" MBS on the SVB books really isn't anything like that CDO plague. In 2008, people were losing faith in the ability to even assign valuations to CDOs; SVB's problem with their MBS portfolio is that the current market valuation is obviously not what they needed it to be just now.
The project had a real mark-to-market agonizing reappraisal in 2010 or so, as reviewed here (from the standpoint of 2013): <https://www.gao.gov/products/gao-13-690t>. Since then development targets have largely been met, and potential customers are much more comfortable with capability and reliability data.
Scathing criticism of the early F-35 development process has always been entirely warranted, and the Pentagon continues to pay the price in rebuilding costs for early "production" airframes. Its also notable that the Navy is being quite, ah, deliberate about F-35C deployment; I think only one carrier is currently rated to operate them. The Marines are more gung-ho about their F-35Bs, but only the F-35A is an export success story so far.
Like all those medical studies that need "in mice" appended to the title, you need to add a "in Canada" for many institutions here. There's no shortage of breathless coverage to go around.
It's like a paradox: Nearly every Canadian can recite a laundry list of grievances they have with Canadian news media but if you actually make any of those accusations, welp, you're not being super-chipper-patriotic enough and get hit with the DVs.
> Canada also has prior non-military liquid-fueled rocket engine development worthy of note:
Non-military... for Canada only, not necessarily for it's allies and partners like China. The Canadian government is known to close it's eyes to dual intent technology transfers to China[0]. Good thing here in America there's ITAR[1] keeping these technologies from the hands of such bad actors.
Canadian universities are also known to host and subsidize degrees for a significant number of Iranian nationals studying aerospace and nuclear engineering, for peaceful purposes I suppose. The Iranian drones currently used in Ukraine by the Russians seem to point toward technology transfers from Canada[2].
> Indeed. This marketing copy is spectacularly over-reaching.
From the tittle and skimming the article I assumed they were claiming to be the first student-led group to reach the Karman Line, completely ignoring USC's record in 2019. Then, I realized they meant with a liquid fuel engine, thus making it a real first (albeit on a technicality). Then actually reading the article I realized the rocket doesn't actually exist and the funding for is hasn't even been secured. They did perform a static test for the engine, but that's not the same as a real flight by any stretch.
Then the article gets really weird, with statements like:
"With the ability to deliver 65 kilograms of scientific payload to over 135 kilometres in space, Starsailor is in the same class as United States suborbital launch companies' rocket-powered vehicles."
Which sounds true except some of these rockets are actually flying. I guess if Starsailor does liftoff and performs to spec it'll be of the same class. Also why aren't they naming any of the companies? Fear of lawsuits?
And then there's the name dropping at the end "The student-led team has scored an impressive number of wins since its inception in 2014, beating other universities such as Stanford, MIT, Caltech and McGill."
I wonder why they went with that list of Universities (to get the right keyword in the article?). I mean, I don't think MIT or Caltech talks about Concordia in it's press releases... And no mention of USC who accomplished a similar goal?
