Forgot your password?

Comment: Re:Model M Keyboard FTW (Score 1) 596

by greg1104 (#46794707) Attached to: Ask Slashdot: What Tech Products Were Built To Last?

Some of the models from 1993 and 1994 have a drainage channel for spills. See the Design section of Model M Keyboard to find out the model numbers. I consider those the peak of the Model M design. The quality dropped noticeably starting in 1995, due to cost cutting changes also mentioned there.

Comment: Re:Model M Keyboard FTW (Score 1) 596

by greg1104 (#46794685) Attached to: Ask Slashdot: What Tech Products Were Built To Last?

Lexmark ruined the design with a set of 1994 cost cutting changes. Models from 1995 and later have a noticeably worse typing feel to them. Unfortunately that lower quality 1995+ version is what Unicomp inherited. They make an OK descendent of the Model M design, but it's surely not the same keyboard as the classic design.

Comment: Re:IBM Model M Keyboard (Score 1) 596

by greg1104 (#46794659) Attached to: Ask Slashdot: What Tech Products Were Built To Last?

The Model M was redesigned by Lexmark in 1994 to use lighter, cheaper components. All of the units from 1995 and later are substandard compared to the earlier ones. The backplate is just one of the problems.

Engineered quality peaked with the 1993 and 1994 models that were updated to have a liquid drainage channel.

Comment: Re:Mikrotik? (Score 0) 96

by greg1104 (#46783925) Attached to: Ask Slashdot: Which Router Firmware For Bandwidth Management?

Amazon is in a death spiral, or is already dead to technical people? Dude, you know how to tell a funny story, I'll give you that. I am more troubled by Bezos's inconsistent stand on patents than knee-jerk characterization of his personal politics. Hint: is he a conservative Republican tool, or a super rich liberal? Labels are so tricky.

Comment: Re:I'd seriously think about a dedicated router (Score 2) 96

by greg1104 (#46783859) Attached to: Ask Slashdot: Which Router Firmware For Bandwidth Management?

I am highly skeptical of claims toward the OSS router firmware scene being less useful than manufacturer provided ones. You're right that speed to support new features lags in OSS, but who cares? I buy the router based on the hardware compatibility list, not the other way around. Reliability and longevity is a lot more important to me than the new shiny. You're also right that today it may be difficult to meet all the requirements with open code, with AC support being a sore point. I'd use that as a reason to delay the purchase until i can though, not as an excuse to head any distance back toward less open development models.

I still have two Linksys WRT54GL units left in operation. Long after Cisco/Linksys stopped worrying about that hardware, I was happily served by the software communities around DD-WRT and then Tomato. Manufacturers like Ubiquiti are useful to me to the extent they embrace that philosophy. In the last year Linksys seems to be moving back in the right direction again. We'll see how that plays out.

I'm also skeptical that having two points of failure in a network can ever be more reliable than one, which complicates your flexibility argument. Whenever I decouple routing and wireless onto separate boxes, problem resolution is harder compared to having a single unit to swap out. One of the reasons I ended up with so many cheap WRT54GL units is that I could easily have a spare with a duplicated configuration for every install. At any scent of trouble, I just replaced the whole unit.

Comment: Re:NONE, get a smart switch (Score 2) 96

by greg1104 (#46783613) Attached to: Ask Slashdot: Which Router Firmware For Bandwidth Management?

I don't know when you got your Netgear GS108T units at, but somewhere in that product's lifecycle it turned bad. My experience mirrors the highest rated critical review at Newegg, circa 2011 and talking about the decline. There are several reasons why the current version of the product only averages 3 stars there, and why 28% of buyers are giving this 1 star now. I have a good, older GS108T and a worthless newer one. Each firmware update is rolling the dice.

That's actually the core argument behind why I won't buy a manufacturer only firmware network product anymore. When the Netgear firmware on a Netgear product is broken and that's the only option, you now have a paperweight. The Tomato firmware upgrade scene for routers is more complicated than I'd like sometimes, but it always gives you multiple options. I'm using an Asus RT-N66 right now, and I don't ever expect its CPU performance is going to be a bottleneck for me. I'm using the Netgear switches only to add more wired ports than it supports.

Comment: Re:OpenSSL OR... (Score 1) 288

by greg1104 (#46761173) Attached to: OpenBSD Team Cleaning Up OpenSSL

It's not "simply" though. The interfaces in OpenSSL and GnuTLS are not swappable APIs. We went over this a few years ago for PostgreSQL, and one of the major issues was having too many OpenSSL-isms in our code to swap easily.

Those of us who dislike bad open source licenses have been trying to kick OpenSSL out of projects for years now, and it hasn't gone anywhere but upward in adoption. I've been amazed at how often I see its advertising clause in the credits of video games I play.

Comment: Re: Customers may benefit... maybe (Score 1) 455

by greg1104 (#46603751) Attached to: Wal-Mart Sues Visa For $5 Billion For Rigging Card Swipe Fees

Coin locked carts have been at US stores before. Here in Baltimore, the BJ's I used to shop at had them in their lot around 5 years ago. It was interesting to watch that unfold as a new social dynamic. I tried to hand over my activated cart with the quarter in it to parents wrangling smaller children when I left.

Comment: Re:Parallel 2/3 didn't work in Windows until 3.3 (Score 1) 196

by greg1104 (#46526719) Attached to: Python 3.4 Released

OK, it took until 3.3 before this was straighforward in Windows. Why is that still relevant today? Python 3.0 came out at the end of 2008, and several parts of the 2.X transition were still pretty rough then. A poster above made a nice comment about how that's played out: "Python 3.3 (or 3.4, as this article is about) is not 3.0 or 3.1. There is a lot of things that have been fixed along the way." Having an upgrade path that's possible to follow smoothly has been a design goal of 3.0 since its early days, but it wasn't quite there yet when 3.0 first shipped. That's history at this point though.

Comment: Re:Time to fork Git? (Score 1) 710

EnterpriseDB is an important part of PostgreSQL development with several contributors, but they still work within the larger development community of contributors. There are other companies with just as many contributors, with one example being how 2ndQuadrant is adding logical replication features.

One way you can tell if an open source project has a real community is whether the project would go on even if the largest company contributing code disappeared. Linux would survive RedHat disappearing, and PostgreSQL would certainly survive EDB going out of business. That's not even a theoretical question, because the PostgreSQL community is informed by having seen it happen once already. A company named Great Bridge hired a good percentage of the PostgreSQL community once, and then failed after running out of VC cash.

Comment: Re:Because... (Score 1) 794

by greg1104 (#46377039) Attached to: Whole Foods: America's Temple of Pseudoscience

Yes, you did miss the expose, there have been hundreds of lawsuits. They went to the supreme court. Farmers who don't use Monsanto's seeds can go out of business from the legal risk they take on. It's a classic protection money racket. You pays your tribute to buy our seeds, or something unfortunate might happen to your crop one day, when our lawyers come to break your kneecaps.

Contemptuous lights flashed flashed across the computer's console. -- Hitchhiker's Guide to the Galaxy