Slashdot Mirror
RIP: Leonard Zubkoff
UnidentifiedCoward writes "LWN.net has a link to a blurb at KTVA, "Alaska State Troopers have recovered the bodies and released the names of two men killed late last week in a helicopter crash in Southeast. They are 38-year-old David Zampino of Fairbanks and 45-year-old Leonard Zubkoff of Crystal Bay, Nevada." Mr. Zubkoff was a linux kernel developer and the maintainer of BusLogic
and DAC960 projects." Leonard was a hell of a nice guy and will be missed.
Leonard will be missed by many. He was Dandelion Digital's sole proprietor. He also has a page about his Linux acheivements, especially his SCSI drivers which are commonly used today.
http://tomgould.com/
According to Dandelions page, Leonard was also interested in the Cryonics movement. While I don't personally think this works / is going to work, I feel sorry that he's not going to get a chance to test it out himself. Unless someone has information to the contrary?
Goodbye Leonard, you will be missed.
Bryn.
There's not much in the FAA report (see below) but it looks like weather wasn't an issue.
-----
IDENTIFICATION
Regis#: 7189T Make/Model: R44 Description: 2000 ROBINSON R-44
Date: 08/29/2002 Time:
Event Type: Accident Highest Injury: Fatal Mid Air: N Missing: N
Damage: Substantial
LOCATION
City: KETCHIKAN State: AK Country: US
DESCRIPTION
2000 ROBINSON 44 HELICOPTER CRASHED IN WINSTANLEY LAKE, LOCATED UPSIDE
DOWN, 2 POB SUFFERED FATAL INJURIES, OTHER CIRCUMSTANCES ARE UNKNOWN,
KETCHIKAN, AK
INJURY DATA Total Fatal: 2
# Crew: 1 Fat: 1 Ser: 0 Min: 0 Unk:
# Pass: 1 Fat: 1 Ser: 0 Min: 0 Unk:
# Grnd: Fat: 0 Ser: 0 Min: 0 Unk:
WEATHER: KTN METAR 08/30/02 0053 UTC 34009KT 10SM SCT040 BKN070
OTHER DATA
Activity: Pleasure Phase: Unknown Operation: General Aviation
Departed: KETCHIKAN, AK Dep Date: 08/28/2002 Dep. Time: 0349
Destination: WINSTANLEY LAKE, AK Flt Plan: NONE Wx Briefing: U
Last Radio Cont: DEPARTING KETCHIKAN
Last Clearance: NONE
FAA FSDO: JUNEAU, AK (AL05) Entry date: 08/30/2002
It was a Robinson R-44. The R-44 is not a kit, and it's far from a toy. Unfortunately, Robinson's have a huge piece of the statistics pie simply because they're the one rotary wing affordable enough to be used for instruction. And students crash a lot.
Most people here probably know more about Leonard Zubkoff, so why don't I talk a little about David Zampino. Mr. Zampino ("Please, call me Dave") was my boss at the company I worked for during the summer between high school and college. At that time, he was working for RCN, a fairly large ISP. He taught me a great portion of what I know about TCP/IP and routing equipment. He was a great boss; on days with lots of customers calling, he would tell us techs (not just 'let' us; he would say, "Be sure to expense that,"!) to expense pizza delivery if we decided to work through lunch. His focus was always, always on having good customer service. He really wanted the company to have a reputation as being dependable and competent, even if it cost time and money. He also did not have much patience for office politics. One time, a customer called and complained that his Unreal server had suddenly become sluggish, and all the players' ping times had gone up 60 miliseconds. I investigated, found a problem in our routing, and sent an email to the appropriate internal mailing list. When the person in charge of the buggy area fired back a reply chastising my inexperience and ridiculous notions, Dave was on it like a hawk. In less than 15 minutes he, too, had investigated the problem, and wrote to the list both to back me up and to castigate the other manager (albeit in a very diplomatic way) for reflexively 'defending his turf' without even looking into the problem!
I don't know a whole lot about him before that time. Mr. Zampino was the founder of Brainstorm (1990-1991 ish), which started as a hardware company making accelerator cards for Macs. They ended up as a local, business-only ISP (long story) which was eventually bought by RCN, which is how he ended up there. While he may not have been a kernel hacker, he was certainly no slouch in Unix operation and programming, nor in hardware design. Although I have not worked for RCN since the summer of 2000, and I believe he left the company earlier this year, I am sure he is fondly remembered by all his co-workers.
Rest in peace, Dave. I am proud to have had the privilege of knowing you.
I wrote about David Zampino in another comment. I neglected to mention in that post that, in addition to being a chip designer, programmer, and IT manager, David Zampino was also known as 'Bat Dave' in my corner of the office because of his ability to fly small planes and helicopters.
Maybe Chris knew Leonard personally and not David. He wouldnt exactly be quallified to comment on someone-he-did'nt-know's personallity any more than you assholes are to be commenting on his.
Too right brother, his work is still doing it's stuff on my box at home
/proc/scsi/BusLogic/0
cat
***** BusLogic SCSI Driver Version 2.1.15 of 17 August 1998 *****
Copyright 1995-1998 by Leonard N. Zubkoff <lnz@dandelion.com>
Configuring BusLogic Model BT-930 PCI Ultra SCSI Host Adapter
Firmware Version: 5.02, I/O Address: 0xDC00, IRQ Channel: 10/Level
PCI Bus: 0, Device: 10, Address: 0xDFFFF000, Host Adapter SCSI ID: 7
Parity Checking: Enabled, Extended Translation: Enabled
Synchronous Negotiation: FUUFFFF#, Wide Negotiation: Disabled
Disconnect/Reconnect: Enabled, Tagged Queuing: Enabled
Driver Queue Depth: 255, Scatter/Gather Limit: 128 segments
Tagged Queue Depth: Automatic, Untagged Queue Depth: 3
Error Recovery Strategy: Default, SCSI Bus Reset: Enabled
SCSI Bus Termination: Enabled, SCAM: Disabled
*** BusLogic BT-930 Initialized Successfully ***
Target 2: Queue Depth 3, Synchronous at 20.0 MB/sec, offset 8
Target 3: Queue Depth 3, Synchronous at 6.67 MB/sec, offset 15
Current Driver Queue Depth: 255
Currently Allocated CCBs: 28
[snip]
For those who don't read METAR (which includes some of us meteorologist), here is the gist:
Temp: 63F/17C
Dewp: 51F/11C
Winds: Northwest at 14mph/12kts
Pressure: 1018.7mb
Sky: mostly scattered or broken between 4000ft/1220m to 7000ft/2135m
Visibility: 10miles/16km
All in all, not a bad day, though it was a little windy. I do agree with an above post that icing COULD be a possibility, but with a surface temp as warm as it was, they would have to be flying pretty high.
Bryan R.
The price of freedom is eternal vigilance, or $12.50 as seen on eBay.....
As an undergrad at CMU, the CS terminal room was rather evenly split between DEC VT52 and Concept C100 terminals. And then there were the "special" terminals: the Concept-LNZ. These amazing little creatures were a result of Leonard's graduate work. They contained custom firmware that the locally hacked version of Unix Emacs contained special support for. It cached frequently displayed tokens in local (off screen) video memory and exchanged an encoded/compressed token stream with the editor. Working over a 2400 bps serial line was an absolute dream on these. It sped up the editing sessions to an amazing degree.
When I asked Leonard in an e-mail if he was "the" LNZ of Concept-LNZ fame, he was rather flabergasted that someone would remember this over 10 years later. He gradually took over support for the Buslogic driver as he was both a better driver writer and had local access to the Buslogic lab to do testing.
When I read this headline, my gut tied in a very tight knot that will not soon be untangled.
We'll miss you, Leonard.
An imperfect plan executed violently is far superior to a perfect plan. -- George Patton
Here are some big runon blurbs I tend to write in order to fondly remember and credit someone.
:} I'd just negotiate with him so that he didn't have to end up scrounging together an engineering team to duplicate all of IT's infrastructure. *wink to mobyone and claw*
:}
I was employee number twentysomething (IT admin) at VA Research, and I was interviewed by Larry, Rob, and Leonard. lnz always had time to randomly consult on the spot with employees in terms of engineering or general technology. Whenever he'd breeze through the office (never coming in before afternoon), ya knew he was kickin some ass. He was often seen smiling. He was one of the first people I was personally aware of to really use Linux itself to make a big dent in major industry, through his work with Buslogic and Mylex SCSI controllers. He told me when Adaptec finally stopped disavowing the existance of Linux, they came to VA and said "We're sorry. Can we play with Linux now?" and lnz said, "No. Too late." He'd already schooled them on Linux from the grassroots on up, forcing them to acknowledge an emerging market. I'm sure he was a strong mentor for that driver engineering and reverse engineering community. Man, that takes devotion and patience.
Ya couldn't mess with his workflow. He had like a mini data center and R&D lab at his house, which he relied solely on at all times, telneting home and xhosting his XEmacs display back to the office when we had public IP addresses for all workstations.
Then with the pre-IPO, he had to move his R&D out of his house into the office. This was when we were in the original garage-like Mountain View office next door to SGI North American sales on Shoreline, and our building's resources was about 3 times overcommitted by our growth rate. We had phone lines and ethernet cords draping out of the ceiling down to shared desks in order to accommodate having new employees per week, and I had to figure out how to route power all around the building using very warm and very illegal 14 gauge extension cords from each available power circuit to wherever in the building lnz's engineers needed them. Routed em like some people route ethernet cables. Such as to lnz's new 1 terabyte file server sitting next to my desk, powered by the women's restroom. That server was lnz's baby; you may have seen it at the March 1999 Linuxworld Expo. He blew that circuit that afternoon. Permanently. The women's restroom never worked again. Thanks to his rapidly growing engineering dept and to our new sales dept, the power generator in the back was hot enough to singe your body hair when you opened the door to it. The fsck alone on that event pushed the ship date back a day or more. Yeah he was shipping 1TB RAID servers with ext2.
lnz inadvertantly taught me a lot about fire and safety codes of Mountain View and Sunnyvale, and he taught me the proper use of the word "cryonics" instead of Hollywood's improper use of "cryogenics". He's one super nice guy. Hope to see ya around, lnz.
The Robinson's suffer from a problem that all lightweight heli's do, and that is low rotor inertia. It makes them very twitchy in flight. Unfortunately, it's not just stundent pilots that crash them. A large number of crashes in R-22's and R-44's happen with experienced pilots at the stick. Just last month a FACTORY Robinson pilot crashed a brand-new R-22 in Texas.
The rotors on a Robinson can actually flex enough to strike the aircraft in flight if you over control them.
Leonard made changes to Emacs on ITS and TOPS-10 when he was at CMU, in order to take advantage of the screen buffer controls (insert line, rather than redraw-rest-of-screen, etc.) on the Heath/Zenith H19/Z19 VT52 clone.
Although as everyone knows Leonard later became a strong contributor to free software, these updates to Emacs he placed under a restrictive license, and vigorously protected his code. RMS was quite upset by this, as were some other folks. Although Stallman's tiff with Symbolics over Lisp Machine source access is often cited as the reason he started the GNU project, I believe that his interactions with Leonard over ZEmacs were an even earlier influence.
So, in some sense, we have Leonard to thank for the Gnu project that he later contributed to.
Here is the earliest Usenet mention that Google has (we weren't all big USENET users in those days -- it was mostly UUCP modem-based systems).
Bye, Leonard. You'll be missed.
There is also a page of memories from some of his friends in the filking community at:
http://www.filking.net/articles/zubkoff.html