You’ve probably seen this ad here. The majority of the video is for another product entirely. It shows a projection from the Sega Toys Homestar, which is $240. (And is amazing. I know; I bought one like 15 years ago.)
You can see it briefly in the video (snapshot below)
The Sega device is amazing, has multiple lenses, and microfilm-style projection discs that create an amazingly lifelike star field. Like, seriously, it’s fucking amazing.
But this advertised device is cheap, and projects a cartoonish scene without regard for realism or focus.
It’s sold by one of a dozen or so cookie-cutter web sites you’ve probably seen running ads on Twitter. Often with nonsense domain names (in the .co TLD) and with most of the sites using the same exact template. Everything they sell is advertised in a misleading way.
I’ve been reporting them, but Twitter doesn’t offer an option to report ads as misleading or fraudulent. I know of many people (some, personally) who’ve ordered seemingly amazing items only to have some totally unrelated junk show up months later. Don’t be fooled.
Incidentally, for more info on that Sega Toys Homestar device, check out their site: segatoys.space/en/index.html
If you’re looking for the most lifelike projection of a clear night sky—the sort you’d see in the dark desert somewhere—the Homestar is absolutely what you want.
(Yes, I know I used “amazing” a lot up there. It’s late, and I’m not in a wordsmithing mood right now. 🤣 Besides, the real Homestar is, indeed, amazing.)
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Docs show the laptop was dropped on Apr 12, 2019. They also show an external drive and its serial number.
Western Digital’s web site says that drive’s **3-year** warranty expires Apr 18, 2022…meaning it was manufactured Apr *18*, 2019.
If the data was on the MBP’s NVMe SSD, it was either readable or not. The TRIM command is constantly shuffling blocks around to level the wear on the memory cells, and any legit data recovery firm will tell you: recovery of damaged/deleted files from an NVME drive is impossible.
The external drive was claimed to be used to recover the data, but there is no “recovery” of a solid state disk drive, such as the one in Hunter’s alleged Mac. If the data is corrupted or deleted, it’s lost forever, unless the Mac’s owner runs a *specific* command from a shell.
My primary isn’t for two weeks. If you’re doing any math around who gets whose Warren’s votes if she drops out before then:
I am a progressive Warren supporter. I decry racial, financial, gender, and social inequality.
I believe that big corporations have an obligation to “pay back” (by way of higher taxes) the rest of us that paved their way.
I believe that free markets result in a consolidation of power and wealth that benefits the already-rich and powerful…which harms of the rest of us.
I believe in civil liberty. The right to privacy. The right to be forgotten. The right to choose. The right to be seen & treated as an equal. The right to an education that improves one’s capacity to contribute to society. The right to not live in fear of guns every-damn-where.
Over ten years ago, @guardian published a story about cults and their charismatic leaders. They included a list of behaviors that should be warning signs of an unsafe group or leader.
I'll list them in this thread, along with sample news stories about Trump that align with them.
1) Absolute authoritarianism without meaningful accountability.
Hitler, Mussolini, and Mao Zedong curated dysfunction in their respective governments, in order to gain power.
Update on Part 2 of my article series discussing #NLP/data science analysis of the Mueller Report:
• new "clean" source of OCRed text found, and will be used for all further articles in the series
• I found an excellent "coreference" tool that I will demonstrate and train
1/6
"coreference": when a noun (e.g. full name) is used once in a paragraph, and that noun is referred to later using another form (e.g. last name, or pronoun). This screenshot from a web-based demo shows how three "coreferences" have been de-referenced to their root.
I have trained this particular tool using the list of references provided at the end of the report, and it works well.
• de-referencing relative dates, and how I did it
• showing only relevant tweets from the correlated timeline using simple text classification tricks
3/6
Mueller Report #NLP analysis, live update: I've started cleaning up the code, and writing my article.
For my non-tech tweeps: the article will include code, but I promise to explain it all! Plus, you'll see the promised timeline of events from the report, tweets, and more!
1/5
For my geek followers: here's what to expect: the article is being written as a Jupyter notebook running with a Python 3.7 kernel. I'll include instructions to set up a venv with a separate kernel, if that's how you roll. I'll also include the requirements.txt.
2/5
If you're using the Anaconda Python distribution (which is awesome), you can run Jupyter in a new conda environment too; instructions will be provided. The notebook and associated artifacts will be posted to GitLab.
3/5
Then: the ethics of using machine learning approaches to guess at what will fit under the redaction boxes.
Short answer: technically feasible (for some definition) with language models that are trained with every indictment, sentencing memo, etc. from Mueller. Ethically wrong.
I’ve been reading a number of posts/pages on the topic, and now that I’m well past the honeymoon portion of this endeavor, the novelty of having a model make even one successful prediction has lost its appeal.