Hacker News Reader

Comment by takeda | Hacker News Reader

I've seen a picture posted on Slashdot how they server rack looks like. There are many very old machines, I am sure that at least one reason is fear that they break during transportation.

Found it: http://www.openbsd.org/images/rack2009.jpg

View on HN

Why not move the machines, and if the Amiga breaks down and they can't find a replacement, end Amiga support? I mean, that's not a wonderful outcome, but what would you prefer to see given the following options?

a) Shut down OpenBSD

b) Shut down Amiga support in OpenBSD

I mean, is it even a hard choice?

Besides, if there are many developers who like developing for Amiga, surely they would be able to find a replacement?

Or hidden option c): ask people to donate money.

Well, yes, clearly. But it's been a month since the initial ask, and this sort of "threat":

the fact is right now, OpenBSD will shut down if we do not have the funding to keep the lights on.

suggests the necessary $20k (cash) has not been forthcoming.

P.S. I understand it's not a threat in the sense of ransom etc, but the most correct word is not coming to my mind.

I completely agree with your association. In my view OpenBSD is very plainly holding itself hostage for $20,000 cash and expecting everyone to accomodate them.

Oh, right, I've seen this movie before. [1]

[1] http://www.youtube.com/watch?v=Z_JOGmXpe5I

It's game theory.

All three options exist, but the maintainers pretending that option b doesn't exist at this point in time increases the probability that option c will succeed.

If the pretense doesn't work, be very much assured that they will go with option b.

The Amiga port isn't live. It hasn't been maintained for a while IIRC.

There are two important points that shouldn't be forgotten about aggressively pushing cross-platform: it retains developers and exposes bugs. There's a great deal of usefulness behind it, beyond simply making it obvious that the workstations we get today are shit.

Ok, Amiga was just an example I pulled off the top of my head.

I am not suggesting all legacy platforms need to be cut. I'm suggesting that it's possibly an acceptable risk, and also if a replacement UltraSPARC simply cannot be sourced, there can't be that many developers working on UltraSPARC anyway. (Just as an example)

That image has been present on the lower right corner of the openbsd homepage for a long time. Given the name of the image, I imagine it's been present since at least since 2009, but my (fuzzy) memory wants me to believe there has been an image of a rack there even before then.

Am I the only one who want's to organize and clean up those racks? :)

The collection of stuff in between the racks is making my neck itch.

Huh, no Sun4/470? I wonder what happened to the one Theo got from me. Too slow I guess but it was a fun machine back in the day.

The AC instructions note made me smile.

It reminded me of a post-it I left in the company lab with a diagram for how to do proper gigabit cross-overs. I could still find it there five years later after they rearranged the lab several times.

I thought most (if not all) gigabit Ethernet implementations auto-detect host-to-host connections and handle the crossover for you.

They do. Maybe he meant 100BASE-T4.

They all do now, but it's still an optional element of the standard.

It was optional in 100Mbit. In gigabit, it is mandatory.

"Automatic MDI/MDI-X Conﬁguration is intended to eliminate the need for crossover cables between simi-lar devices. Implementation of an automatic MDI/MDI-X conﬁguration is optional for 1000BASE-T devices. The assignment of pin-outs for a 1000BASE-T crossover function cable is shown in Table 40-12 in 40.8."

Here's a link that does not require IEEEXplore access: http://www.scribd.com/doc/46141801/802-3ab-1999

I say, I stand corrected. I was sure I read the opposite somewhere, but I would not dream of contesting your reference.

There must be more to it than that. Two racks aren't using $20k in electricity.

Reply on news.ycombinator.com