New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Now on LEB: Interview with jsg
raindog308
Administrator, Veteran
in General
@jsg originally resisted doing an interview because he thought no one would be interested, but I think if you give it a read you'll find his perspective on security, static analysis of code, chips, and BSDs to be quality reading. Enjoy!
https://lowendbox.com/blog/interview-qa-with-lowendtalk-member-jsg/
Thanked by 14seriesn poisson Neoon cociu simonindia Pwner muffin AlwaysSkint chocolateshirt jsg dfroe Tr33n vedran limited
Comments
Well most of the stuff @jsg talks about is foreign to me. I'm glad someone in our midst is a pro in security and have the experience to smell the usual bullshit with security hundredths of miles away. Learned a lot and I can see myself spending the next decade trying to at least know how to approach security in the level of JSG.
@raindog308 This sort interviews really expose the best in LET. So thank you and others who make this happen including @jsg for spending his valuable time on this one.
An interesting (and naturally long) read, during morning coffee.
Cheers!
Certainly helps to understand some of the @jsg comments.
Yes. I also read it during my coffee. Once you get into a person's way of thinking, you can understand many of their long comments and ideas shared around here.
@jsg is old-school, seeing computing through it's core, while extrapolating technology through time and security as they evolved. I can not possibly know how right or wrong he is, because I do not even know those programming or security languages he mentioned, but his different approach into thinking definitely nourishes my respect for him.
This interview made me enjoy my coffee better. Thank you @jsg and @raindog308 .
Good stuff man! One of the very few interviews, that I enjoyed reading. Thanks @raindog308 @jsg
My compliments to @raindog308 as it looks like more than 3 people read the interview, so it seems he was right to (very gently) pushing me into it.
And thanks also to the patient readers for their interest in a person I still consider not particularly interesting (yes, I can be a bit stubborn).
Whatever, my main two reasons to do as asked were (a) to contribute my bit to LET, and (b) to provide a glimpse into a field that I consider important and very interesting. And compliments again to raindog308 who guided me so well with his questions and to @jbiloh who invested way more than money to "renovate" LET and to make it great again!
Thank you
Well worth reading. Thanks to both @jsg and @raindog308 for investing the time into putting something out that was interesting and detailed.
I also started in Electronics, when the local employment in that area was just about to collapse - the days of Burroughs, Hughes etc. At least I got to spend some years in a High Voltage lab. at a large industrial site, that could be 'fun'. Though used extensively in the same workplace, I never got into the microcontroller side of things. (Apart from blowing up a BBC Micro interface when trying to control a variac!)
Moved over to Application/Hardware Support, with a few other diversions along the way before doing this hosting/web melarkey, to bring in some pocket money.
Nowhere near as in-depth as jsg's exploits.
@AlwaysSkint
Haha, my first work with "microprocessors" was without an MCU. In that job they threw CMOS (primitive logic gates) chips at me and I had to somehow come up with a digital signal via a PC card I had to design that grabbed a purely analog signal from some high-tech (well, back then) measurement device that was basically something like a slow oscilloscope (about 2 MHz).
When I was a little shocked and asked why I was told that the affordable MCUs couldn't handle the required speed and those which could were too expensive (basically high end MCUs with on-die DSP). It took quite a few years before I worked with a real MCU and that was a private hobby project because meanwhile my job had become software.
Fun fact: that project somehow stuck with me and I still do quite a bit of work (hobby only) on dual slope 5.5, 6.5 and even a 7 digits DMMs (DC, dual slope) and algorithmic MCU DMMs (AC) with quite decent accuracy (significantly better than most 1k+$ DMMs on the market). But now I better stop because DMMs are a topic that gets me quite quickly started ... *g
Getting horned up by DMM's might be the creepiest thing I've read on LET, at least for a while.
I bet the v2 benchmark software will drop on the 32nd of never.
And much of the AMD/Intel thoughts can better be understood reading websites that follow AMD and Intel, than his random conspiracy thoughts. Saying AMD had a long time and Intel didn't is opposite of reality (the AMD turnaround story was publicized years ago when Ryzen launched). AMD fucked up royally with bulldozer and quickly came out with their fastest design to market ever at the time. They've been beating their own targets for several years now. Intel, missed years and years of roadmap targets. Intel's R&D budget dwarfs AMD, so one can easily argue AMD "is (intellectually and engineering-wise) better than intel".
I have to say, was really an interesting read. Thanks.
As for AMD and being better than Intel engineering wise, Id have to agree and disagree with that. We all saw what AMD did without Keller. Bulldozer (and forward before Zen) was a disaster even tho I bought it all the way from 8150 to 9590. Thankfully even AMD itself understood they needed to bring back Keller to get out of that hole. Im not trying to imply that other engineers in AMD are dumb or something like that but it really seems they are not on the level of Jim either. moar cores!
Altho, their graphics part of the business, god damn leave gcn behind already.
Possibly, though I see the attraction; high persistence analogue (non-storage) 'scopes stir my interest.
As for Intel/AMD; well a monopoly is never good and it's a shame Cyrix (and others) couldn't make a bigger impact.
He is much more polite on interview