home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   alt.os.development      Operating system development chatter      4,255 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 2,358 of 4,255   
   Rod Pemberton to Grant Taylor   
   Re: TLS 1.0 (1/2)   
   22 Jun 21 03:52:14   
   
   From: noemail@basdxcqvbe.com   
      
   On Thu, 17 Jun 2021 22:29:54 -0600   
   Grant Taylor  wrote:   
   > On 6/17/21 3:48 PM, muta...@gmail.com wrote:   
      
   > > [...]   
   > I would rather you contribute to [...]   
      
   This strikes me as rather wrong.   
      
   I mean, I would rather lots of   
   other people do lots of other   
   things, which would benefit me.   
   Of course, I'm not entitled.   
      
   > > The trouble is that the system I am most interested in is PDOS/386   
   > > which only supports C90. Well, sort of.   
   >   
   > Either it does, or it does not.   
      
   If it only supports a subset of C90,   
   then it both does and does not.   
      
   It "does" support C90 /only/.   
   It "doesn't" support C90 /fully/.   
      
   Adverbs.  Missing.   
      
   > How is an operating system the limiting factor of what level of C is   
   > supported?   
      
   I'd say the OS implementer is "the limiting factor of   
   what level of C is supported" in an OS.   
      
   I.e., the designer may choose to limit what portions   
   of C and the C library in used to implement the OS.   
      
   E.g., I limited my OS project to the C language proper,   
   and a few string and memory functions, as they were   
   OS independent for the C compilers I was using.  I've   
   limited many utilities that I wish to bootstrap to just   
   the C language proper, stdio.h, stdlib.h, and string.h.   
      
   >  Shouldn't that be the compiler's job?   
   >   
      
   So, it's the compiler's job to determine what level   
   of C is used in an OS?  Just how much "intelligence"   
   do you expect a compiler to be imbued with?  How do   
   you expect humans to enhance the C compiler with that   
   much intelligence without using Artificial Intelligence? ...   
   The C compiler would need to be nearly sentient.   
      
   > > I want to flesh out what is possible with C90.   
   >   
   > You're free to impose your own artificial restrictions.   
      
   Does he really need you to tell him that?   
      
   He seems to fully understand he has gone   
   off-into-the-weeds, unlike other people   
   on Usenet.  Perhaps, it's brainstorming,   
   or urgency, and not insanity.   
      
   > > What existing C90 browser do you suggest I use instead of creating   
   > > my own? Preferably one that is public domain.   
   >   
   > I have no idea what falls within your artificial self imposed   
   > restriction.   
      
   Nothing.   
      
   C90. Maybe.   
   Public domain?  Nothing.   
      
   > I'm quite certain that there's source code for Netscape from the '90s   
   > available on various Linux archives.  I'm confident that there is   
   > source code for other browsers.   
      
   Am I correct in saying that your intent seems   
   to be to redirect Paul from his OS project to   
   some other huge time-wasting project? ...   
      
   > > Moving OpenSSL code to the modem means that my OS is uncomplicated.   
   >   
   > You're committing multiple layering violations for a very   
   > questionable idea.   
      
   What is a "layering violation"?  Bad hair cut?   
      
   Over any physical connection there can only be   
   a single stream of data, regardless of serial,   
   parallel, or multiplexed, etc.  If you wish to   
   subdivide that data and call various pieces of   
   it different things, e.g., header, signaling,   
   data, checksum, it doesn't change the fact that   
   there is only one stream of data over the   
   physical connection.  In other words, the data   
   is all merged together in some format.  You   
   can call it layers, if you wish, or because   
   some protocols specifies it that way ...   
      
   > > All I want is my modem back. I never agreed to give it up.  Or the   
   > > attached BBS software.   
   >   
   > What do you mean "want is my modem back"?  I don't recall anyone   
   > taking your modem from you.   
      
   Instead of chastising him for his preference of   
   the ancient and IBM, you could've had him search   
   for a "USB modem".  Dial-up. 56K. Fax modem.   
   They're still being manufactured and sold, e.g.,   
   e-mail for laptops on business trips.  He could   
   buy one tomorrow.  Unfortunately, an "Ethernet   
   modem" is a cable modem, and isn't dial-up, but   
   there are Ethernet-to-RS232C adapters which he   
   could use with an external dial-up modem.   
      
   > Similarly, your idea of moving OpenSSL into the modem, how are you   
   > going to communicate all of the parameters that are used to establish   
   > SSL / TLS connections to the modem?   
      
   That's easy.  Webpages.   
   Just like an Ethernet cable modem.   
      
   > How are you going to provide the   
   > trusted root certificate store?   
      
   Via any file transfer app or protocol? ...   
      
   > Modem's don't have any storage to   
   > speak of (NVRAM is miniscule) and /etc/ssl on my modern system is   
   > about half a MB.  How do you manage that certificate store on the   
   > modem?   
      
   Since it's apparently only in his head   
   at this point, it will ...   
      
   Or, you can think of a cable modem, which   
   supports many protocols, is frequently   
   implemented as an embedded Linux computer,   
   and has enough storage to update it's   
   software.  Except, he wants dial-up ...   
      
   > How does the OpenSSL on the modem initiate a TCP/IP connection when   
   > modem's don't speak TCP/IP.  Or are you throwing that in too?   
      
   SLIP.  PPP.   
      
   > A dial-up PSTN based modem can't really go any faster than about ~30   
   > kbps (line rate) because of problems encoding data.  Problems as in   
   > you need to sample the line at least twice as fast as the signal you   
   > want to send through it to have any hope of getting things to work.   
   > The 33.6 ~ 56 kbps (serial rate) was data compression applied to the   
   > serial stream to shrink it down to the ~30 kbps (line rate).  So ...   
   > how are you going to make a dial up modem faster than physics can   
   > allow?   
      
   They did improve the transmission rate   
   somewhat, but I don't recall as to what,   
   as dial-up modems were basically obsolete   
   by then.   
      
   > Even if you're talking Ethernet / {DSL,DOCSIS}, you are still limited   
   > to what the {DSL,DOCSIS} network can carry.   
   >   
   > So ... what is going to be faster about your modem?   
      
   Static ram?  ...   
      
   Asynchronous clocking? ...   
      
   Memory? ...   
      
   Embedded microprocessor?  I.e., not a 1MHz 8-bit   
   6502 or Z80 etc.  E.g., a modern GHz class such   
   as an ARM or x86.   
      
   I'd suspect the last two, at least.   
      
   > > But all this should be transparent. At least if I follow the rules,   
   > > and even if the rules are only apparent in hindsight. Namely:   
   > >   
   > > 1. Apps should fopen COM1 or preferably a user-defined string to   
   > > get access to the modem,...   
   >   
   > What might the user defined string be?  Or more importantly, what   
   > might it represent?  What significance does what it represents have?   
      
   Magic number.  Always best if hard-coded ...   
   The deeper it's buried inside the code base,   
   the better.  The point is to make sure   
   that absolutely no one can ever find it.   
   If the #1 rule of real estate is "Location,   
   Location, Location," then the #1 rule of   
   programming is "Obfuscation, Obfuscation,   
   Obfuscation."   
      
   (I'm marking that as sarcasm and humor,   
   as someone, at some point, is unlikely   
   to grasp this fact ...  And, I'll get   
   ripped for being dead serious and wrong.)   
      
   > > ...so let the OS take care of that.   
   >   
   > Wait a minute.  You have been saying that you wanted the modem to   
   > take care of this.  Now you are wanting the OS to take care of this.   
   > Which is it?  If it's now the OS and no longer the modem, why the   
   > change?   
      
   ...   
      
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca