Browsed by
Month: November 2024

The ‘Home Forward’ experiment is going badly

The ‘Home Forward’ experiment is going badly

“There is one girl here who they are trying to get her out of here. She has tried to stab me twice and I filed reports on that”

“The problems with this building happen to be TPI moving in drug addicts who are not rehabilitated, and need to be rehabilitated, but they are not.”

In the past month, there have been two homicides in front of the the Home Forward Louisa Flowers building in the Pearl – and an egregious animal abuse case caught on camera in their elevator. The building has graffiti on walls and windows. Residents report apartments that smell of fentanyl, unwanted guests are often found in the building, and residents shared multiple photos from inside of trash piling up in the hallways and stairwells.

The response?

On Thursday the building leadership had a pizza party for all residents where they invited case workers and other community resources to connect with the residents. Home Forward workers said they hope to ‘foster strong, supportive networks among the residents,’

Sounds like exactly like the kind of comically ineffective plans we’ve come to expect when dealing with homicidal residents, serious drug addiction, and a quickly deteriorating living environments in Portland. Instead of experts coming in to deal with serious drug addiction and violence – we’ll have a pizza party and talk about it. Meanwhile, animals are being abused openly, fentanyl is wafting around the building, and other residents are having their lives threatened on a daily basis.

Sadly, this is what happens time and again when largely untrained non-profit employees (non-profits who are switched out every time the money runs out) try to deal with very serious criminal and drug abuse issues in Portland.

Articles:

Portland Mayor Ranked Choice Voting visualized

Portland Mayor Ranked Choice Voting visualized

Portland just tried ranked choice voting. It was an interesting experience. I’m not sure if I’m 100% sold just yet, but it seems to have worked reasonably well. If nothing else – it’s a fascinating dive into the data.
On the plus side: I did like having the ability to pick 2nd and 3rd choice candidates. In one case, my 2nd choice candidate won.
On the down side: it required a LOT more work. There were almost 20 candidates for mayor alone, and a decent number of them didn’t submit any information about themselves, were odd-ball one-issue candidates, or were borderline quacks. This easily took me 2-4x the time I would have normally spent. Doing this for a dozen candidates at state and local levels would be exhausting.

As with all things, the unintended consequences are likely what is most interesting. I suspect it’s going encourage candidates to start overlapping on stances in order to steal/appeal each other’s 2nd and 3rd votes – especially if they are not a front runner. This could make the voter’s work even more tough as it’s more like splitting hairs than decisive differences. I think it’s also going to encourage candidates to be more homogenous. Outliers and more extreme ends – on both the left and right – were very soundly defeated. This is probably a good thing in such a far left state like Oregon that’s had some pretty extreme candidates in previous elections. It’s definitely going to make campaign strategy much more interesting, and likely break up the entrenched homogenous political structure of Portland.

Anyway, if you’re curious how to see how each round of voting went, the Multnomah website has a neat visualization how each of the rounds worked out:

Alternatively, a local forum user put together a Sankey diagram of the way the votes flowed from one candidate to the next as candidates were eliminated:

The district 4 councilor race was even more crazy:

Here was district 2 with Kanal starting out a resounding 3rd and ending up bubbling to the top, while Guiney started first and ended 2nd:

Bonus:

KGW8 did a great mayoral candidate interview. I liked the format a LOT. They had some pre-canned questions, but I loved the fact they asked audience submitted ‘raise your hand’ questions that made the candidates actually state their opinions in a yes/no fashion instead of just waffling around like the career candidates usually do.

Programming for the Larrabee/Xeon Phi

Programming for the Larrabee/Xeon Phi

Back in the day, I worked on this little project called Larrabee – which later turned into the Intel Xeon Phi coprocessor. It was an ambitious and exciting platform. It consisted of a ton of 512 bit wide instructions to operate like a lot of streaming GPU architectures, yet was fully general purpose x86.

It turned out that getting performance out of this hardware was difficult. In order to get the full potential of the hardware, you simply had to utilize the vector units. Without that, it is like writing a single threaded app on a 8 core system. Single SIMD lane operation just wasn’t going to cut it as was written about in 2017 International Journal of Parallel Programming article:

“Our results show that, although the Xeon Phi delivers a relatively good speedup in comparison with a shared-memory architecture in terms of scalability, the relatively low computing power of its computational units when specific vectorization and SIMD instructions are not fully exploited makes this first generation of Xeon Phi architectures not competitive”

Using the Xeon Phi Platform to Run Speculatively Parallelized Codes

The paper, and the host of others linked on the page as references, are a good read and gives some hints why fixed-function GPUs have an advantage when it comes to raw streaming throughput. Hint: cache and data flow behavior is as, if not more, important as utilizing vectorization in such architectures.

It just crawled out of the woodwork…

It just crawled out of the woodwork…

I loved the Twilight Zone as a kid. Every year one of the local stations would put on a Twilight Zone marathon and play episodes back to back, 24×7, for the entire memorial day weekend. I used to try staying up all night to watch every episode. I usually made it about 18-20 hours before finally drifting off on the couch.

The Outer Limits was also known for tales of the strange and unusual, but I found they weren’t nearly as compelling (If I’m honest, half of the Twilight Zone episodes weren’t really compelling either). But recently I ran across this episode titled ‘It Crawled Out Of The Woodwork‘ and I have to say that it’s incredible. The story could easily be made into a full-length movie and be pretty terrifying. The antagonist was frightening – and even for it’s era.

The story is based on a great question – likely based on the strange and frightening discoveries occurring during the nuclear age. Namely, while delving into the secret depths of the universe, we may find or unleash terrifying things. Even more, those things might just ‘crawl out of the woodwork’ of our experiments in unexpected and uncontrollable ways.

Definitely worth a watch.

Neat visualization

Neat visualization

This could make a fun little demo – flying through a cityscape with buildings that are constantly generated by AI – getting funkier and funkier as you go along