SMOLNET PORTAL home about changes
[root][parent][raw file]
tgtimes6.txt - tgtimes - The Gopher Times
(URL) git clone git://bitreich.org/tgtimes git://enlrupgkhuxnvlhsf6lc3fziv5h2hhfrinws65d7roiv6bfj7d652fid.onion/tgtimes (git://bitreich.org)
(DIR) Log
(DIR) Files
(DIR) Refs
(DIR) Tags
(DIR) README
---
tgtimes6.txt (32733B)
---
1
2
3
4 The Gopher Times
5
6 ____________________________________________________________
7
8 Opus 6 - Gopher news and more - Oct. 2022
9 ____________________________________________________________
10
11
12
13
14 Sentient Regex tgtimes
15 ____________________________________________________________
16
17 Can there be a sed one-liner that implements Artifi-
18 cial Intelligence? Depending on how you define Arti-
19 ficial Intelligence, it may!
20
21
22 sed -r 's/Is ([^y]*)?/Absolutely, (1)./
23 s/Is (.*y.*)?/I do not think that (1)./'
24
25 How does it work for you? How more accurate than this
26 is machine learning going to become to answer our ex-
27 istential questions?
28
29
30
31 fold, fmt, par: get your text in order katolaz
32 ____________________________________________________________
33
34 If you happen to read plain text files (e.g., phlog
35 posts), you have probably noticed that, especially on
36 gopher, the lines of a text file tend to be wrapped
37 all to a similar length. Some authors are very strict
38 on the matter, and like all the lines to be "justi-
39 fied" (i.e., all adjusted to have exactly the same
40 length, by inserting a few spaces to get the count
41 right). Some other authors (including myself) just do
42 not allow any line to be longer than a certain amount
43 of characters (in this case, as you might have no-
44 ticed, the magic number is 72). But how to they manage
45 to do that?
46
47 Most common editors have a command to format a para-
48 graph ('M-q' in Emacs, 'gwip' or '{gq}' in vim normal
49 mode, etc.). But obviously, there are several Unix
50 tools that can help you getting the right formatting
51 for your files. We are talking of fold(1), fmt(1), and
52 par(1), so keep reading if you want to know more.
53
54 The oldest one is probably fold(1) (and it is also the
55 only one to be defined in the POSIX standard...). It
56 will just break each line to make it fit a given
57 length in characters (by default, 72, which is indeed
58 a magic number). Let's see how to wrap the lines of
59 this post at 54 characters:
60 ____________________________________________________________
61
62 $ fold -w 54 20190213_fold.txt | head -10
63 fold, fmt, par: get your text in order
64 ============================================
65 If you happen to read plain text files (e.g., phlog po
66 sts), you have
67 probably noticed that, especially on gopher, the lines
68 of a text file
69 tend to be wrapped all to a similar length. Some autho
70 rs are very strict
71 on the matter, and like all the lines to be "justified
72 $
73 ____________________________________________________________
74
75 Notice that fold(1) did not really think twice before
76 breaking "posts" or "authors" across two lines. This
77 is pretty inconvenient, to say the least. You can ac-
78 tually force fold(1) to break stuff at blank spaces,
79 using the '-s' option:
80 ____________________________________________________________
81
82 $ fold -w 54 -s 20190213_fold.txt |head -10
83 fold, fmt, par: get your text in order
84 ============================================
85
86 If you happen to read plain text files (e.g., phlog
87 posts), you have
88 probably noticed that, especially on gopher, the
89 lines of a text file
90 tend to be wrapped all to a similar length. Some
91 authors are very strict
92 on the matter, and like all the lines to be
93 $
94 ____________________________________________________________
95
96 Nevertheless, the output of fold(1) is still quite
97 off: it breaks lines at spaces, but it does not "join"
98 broken lines to have a more consistent formatting.
99 This is where fmt(1) jumps in:
100 ____________________________________________________________
101
102 $ fmt -w 54 20190213_fold.txt |head -10
103 fold, fmt, par: get your text in order
104 ============================================
105
106 If you happen to read plain text files (e.g., phlog
107 posts), you have probably noticed that, especially on
108 gopher, the lines of a text file tend to be wrapped
109 all to a similar length. Some authors are very strict
110 on the matter, and like all the lines to be
111 "justified" (i.e., all adjusted to have exactly the
112 same length, by inserting a few spaces to get the
113 $
114 ____________________________________________________________
115
116 Now we are talking: fmt(1) seems to be able to to "the
117 right thing" without much effort, and it has a few
118 other interesting options as well. Just have a look
119 at the manpage. Simple and clear.
120
121 Last but not least, par(1) can do whatever fmt(1) and
122 fold(1) can do, plus much, much more. For instance:
123 ____________________________________________________________
124
125 $ par 54 < 20190213_fold.txt | head -10
126 fold, fmt, par: get your text in order
127 ============================================
128
129 If you happen to read plain text files (e.g., phlog
130 posts), you have probably noticed that, especially on
131 gopher, the lines of a text file tend to be wrapped
132 all to a similar length. Some authors are very
133 strict on the matter, and like all the lines to be
134 "justified" (i.e., all adjusted to have exactly the
135 same length, by inserting a few spaces to get the
136 $
137 ____________________________________________________________
138
139 will give more or less the same output as fmt(1). But:
140 ____________________________________________________________
141
142 $ par 54j < 20190213_fold.txt | head -10
143 fold, fmt, par: get your text in order
144 ============================================
145
146 If you happen to read plain text files (e.g., phlog
147 posts), you have probably noticed that, especially on
148 gopher, the lines of a text file tend to be wrapped
149 all to a similar length. Some authors are very
150 strict on the matter, and like all the lines to be
151 "justified" (i.e., all adjusted to have exactly the
152 same length, by inserting a few spaces to get the
153 $
154 ____________________________________________________________
155
156 will additionally "justify" your lines to the pre-
157 scribed width, while: something like:
158 ____________________________________________________________
159
160 $ head file.h
161 *
162 * include/linux/memory.h - generic memory definition
163 *
164 * This is mainly for topological representation. We define the
165 * basic "struct memory_block" here, which can be embedded in per-arch
166 * definitions or NUMA information.
167 *
168 * Basic handling of the devices is done in drivers/base/memory.c
169 * and system devices are handled in drivers/base/sys.c.
170 *
171 $
172 ____________________________________________________________
173
174 can be easily transformed into:
175 ____________________________________________________________
176
177 $ par 40j < file.h
178 *
179 * include/linux/memory.h - generic
180 *memory definition
181 *
182 * This is mainly for topological
183 * representation. We define the basic
184 * "struct memory_block" here, which can
185 * be embedded in per-arch definitions
186 * or NUMA information.
187 *
188 * Basic handling of the devices is
189 * done in drivers/base/memory.c and
190 * system devices are handled in
191 * drivers/base/sys.c.
192 *
193 * Memory block are exported via
194 * sysfs in the class/memory/devices/
195 * directory.
196 *
197 *
198 $
199 ____________________________________________________________
200
201 Pretty neat, right?
202
203 To be honest, par is not the typical example of a unix
204 tool that "does exactly one thing", but it certainly
205 "does it very well" all the things it does. The author
206 of par(1) felt the need to apologise in the manpage
207 about the style of his code and documentation, but I
208 still think par(1) is an awesome tool nevertheless.
209
210
211 fold(1) appeared in BSD1 (1978-1979)
212
213 fmt(1) appeared in BSD1 (1978-1979)
214
215 par(1) was developed by Adam Costello in 1993, as a
216 replacement for fmt(1).
217
218
219
220
221 GNU tar(1) extraction is quadratic tgtimes
222 ____________________________________________________________
223
224 When implementing something from the ground, it gets
225 possible to build-up a simple home-baked file format
226 or protocol looking perfect without any cruft and
227 legacy. Easy to implement, fast to adopt, supporting
228 everything you need from it, and not much more...
229 Likely an alternative to a huge elephant in the room:
230 the current standard in place used by everyone, huge,
231 with many extensions with many use-cases...
232
233 Why bother, then, with implementing the huge and dif-
234 ficult file format or protocol? Maybe because it
235 would be used by many software, and writing data in
236 this slightly more bloated format would help making it
237 compatible with all the software that already support
238 it.
239
240 In this compromise, a limit can be drawn, across which
241 the big and bloated format or protocol is dropped in
242 favor of a simpler, more reasonable, less time-wasting
243 alternative, eventually home-brewed.
244
245
246 The result is a new tar implementation written for the
247 single special-case of a 1.1 TiB file! [1]
248 1 https://mort.coffee/home/tar/
249
250
251
252
253 BYTE Magazine Covers tgtimes
254 ____________________________________________________________
255
256 The BYTE magazine lives among the legends of computer
257 magazines.
258
259 Being a paper glossy magazine, it had fancy covers.
260 Our usual data archivist heroes, Archive.org, have a
261 large collections of covers for these things. [1]
262
263 On another level of effort, someone with passion and
264 patience, actually went through recreatinhg the scene
265 coming from these covers, that never really existed...
266 Until they did! [2]
267
268 >> In the 1970s and 1980s, Byte magazine featured cov-
269 ers with beautiful, surreal paintings by Robert F.
270 Tinney. What if the scenes that Mr. Tinney imagined
271 actually existed in real life? And what if, as Mr.
272 Tinney was painting them, there was a photographer
273 standing next to him, capturing the scene on film?
274
275 >> That's the idea behind this site. I created and
276 photographed real-world objects and composited the
277 images together in order to show what Mr. Tinney's
278 images might look like in real life.
279 1 https://archive.org/details/byte-magazine
280
281 2 https://bytecovers.com/
282
283
284
285 An experiment to test GitHub Copilot's legality seirdy
286 ____________________________________________________________
287
288 >> This article was posted on 2022-07-01 by Rohan Ku-
289 mar [1] and is now republished on this newspaper,
290 with permission (CC-BY-SA 4.0).
291
292 Preface
293
294 I am not a lawyer. This post is satirical commentary
295 on:
296
297 o The absurdity of Microsoft and OpenAI's legal justi-
298 fication for GitHub Copilot.
299
300 o The oversimplifications people use to argue against
301 GitHub Copilot (I don't like it when people agree
302 with me for the wrong reasons).
303
304 o The relationship between capital and legal outcomes.
305
306 o How civil cases seem like sporting events where peo-
307 ple “win” or “lose”, rather than opportunities to
308 improve our understanding of law.
309
310 In the process, I intentionally misrepresent how the
311 judicial system works: I portray the system the way
312 people like to imagine it works. Please don't make
313 any important legal decisions based on anything I say.
314
315 The only section you should take seriously is “Con-
316 text: the relevant technologies”.
317
318 Introduction
319
320 GitHub is enabling copyleft violation at scale with
321 Copilot. GitHub Copilot encourages people to make
322 derivative works of source code without complying with
323 the original code's license. This facilitates the
324 creation of permissively-licensed or proprietary
325 derivatives of copyleft code.
326
327 Unfortunately, challenging Microsoft (GitHub's parent
328 company) in court is a bad idea: their legal budget
329 probably ensures their victory, and they likely al-
330 ready have a comprehensive defense planned. How can
331 we determine Copilot's legality on a level playing
332 field? We can create legal precedent that they haven't
333 had a chance to study yet!
334
335 A chat with Matt Campbell about a speech synthesizer
336 gave me a horrible idea. I think I know a way to find
337 out if GitHub Copilot is legal: we could use its legal
338 justification against another software project with a
339 smaller legal budget. Specifically, against a speech
340 synthesizer. The outcome of our actions could set a
341 legal precedent to determine the legality of Copilot.
342
343 Context: the relevant technologies Let's cover the
344 technologies and actors at play before I start my evil
345 monologue.
346
347 Exhibit A: GitHub Copilot
348
349 GitHub Copilot is a predictive autocompletion service
350 for writing software. It's powered by OpenAI Codex,
351 [2] a language model based on GPT-3. [3] It was
352 trained using the source code of public repositories
353 hosted on GitHub, regardless of their licensing. In
354 response to a Request for Comments from the US Patent
355 and Trademark Office, OpenAI claimed that “Artificial
356 Intelligence Innovation”, such as code written by
357 GitHub Copilot, should be considered “fair use”. [4]
358
359 Many of the code snippets it suggests are exact copies
360 of source code from various GitHub repositories. For
361 an example, see this tweet: I don't want to say any-
362 thing but that's not the right license Mr Copilot.
363 [5] by Armin Ronacher [6] It contains a screen record-
364 ing of Copilot suggesting this Quake code. [7] When
365 prompted to do so, it obediently fills in a permissive
366 license. That permissive license violates the Quake
367 code's GPL-2.0 license. Copilot provides no indica-
368 tion that a license violation is taking place.
369
370 GitHub performed its own research into the matter.
371 [8] You can read about it on their blog: GitHub Copi-
372 lot research recitation, [9] by Albert Ziegler. [10]
373 I'm not convinced that it accounts for the fact that
374 suggested code might have mechanical alterations to
375 match surrounding text, while still remaining close
376 enough to trained data to be a license violation.
377
378 Exhibit B: The Eloquence speech synthesizer
379
380 I recently had a chat with Matt on IRC about screen
381 readers and different types of speech synthesizers. I
382 mentioned that while I do like some variety, I always
383 find myself returning to the underrated robotic voice
384 of eSpeak NG. [11] He shared some of my fondness, and
385 also shared his preference for a similar speech syn-
386 thesizer called Eloquence.
387
388 Downloads of Eloquence are easy to find (it's even in-
389 cluded with the JAWS screen reader), but I struggle to
390 find any “official” pages about the original Elo-
391 quence. Nuance acquired Eloquent Technology, the de-
392 veloper of Eloquence. Microsoft later acquired Nu-
393 ance.
394
395 Eloquence sample audio
396
397 Matt recorded this sample audio clip of Eloquence
398 reading some text. [12] The text is from the intro-
399 duction of Best practices for inclusive textual web-
400 sites. [13]
401
402 >> My primary focus is inclusive design. Specifi-
403 cally, I focus on supporting underrepresented ways to
404 read a page. Not all users load a page in a common
405 web-browser and navigate effortlessly with their eyes
406 and hands. Authors often neglect people who read
407 through accessibility tools, tiny viewports, machine
408 translators, “reading mode” implementations, the Tor
409 network, printouts, hostile networks, and uncommon
410 browsers, to name a few. I list more niches in the
411 conclusion. Compatibility with so many niches sounds
412 far more daunting than it really is: if you only se-
413 lectively override browser defaults and use plain-
414 old, semantic HTML (POSH), you've done half of the
415 work already.
416
417 I like the Eloquence speech synthesizer. It sounds
418 similar to the robotic yet predictable voice of my
419 beloved eSpeak NG, but with improved overall quality.
420 Unfortunately, Eloquence is proprietary.
421
422 Exhibit C: Deep learning speech synthesis
423
424 Deep learning speech synthesis [14] is a recent ap-
425 proach to speech synthesizer creation. It involves
426 training a deep neural network on voice samples, and
427 using the trained model to generate speech similar to
428 a real human voice. One synthesizer using deep learn-
429 ing speech synthesis is Mozilla's TTS. [15]
430
431 Zero-shot approaches could allow a pre-trained model
432 to generate multiple different voices. YourTTS [16]
433 is one such example. This could allow us to syntheti-
434 cally re-create a person's voice more easily.
435
436 My horrible plan
437
438 My horrible plan revolves around going through two
439 different lawsuits to set some judicial precedents;
440 these precedents could improve the odds of succeeding
441 in a lawsuit against Microsoft for Copilot's licensing
442 violations.
443
444 If this succeeds, we have new legal justification that
445 GitHub Copilot is illegal; if it fails, we have still
446 gained a means to legally re-create proprietary soft-
447 ware. It's a win-win situation.
448
449 Part One: set a precedent
450
451 1. Train a modern text-to-speech (TTS) engine using
452 the voice a proprietary one made by a company with a
453 small legal budget. Keep the model's internals hid-
454 den.
455
456 2. Then release the final TTS under a permissive li-
457 cense. Remember, we're still keeping the machine-
458 learning model hidden!
459
460 3. Wait for that company to file suit. [17]
461
462 4. Win or lose the case.
463
464 Part Two: use that precedent against Microsoft's Nu-
465 ance
466
467 Our goal here is to get the same legal outcome as the
468 low-stakes “trial run” of Part One.
469
470 Microsoft owns Nuance. Nuance previously bought Elo-
471 quent Technology, the developers of the Eloquence
472 speech synthesizer.
473
474 1. Repeat Part One against Nuance speech synthesizers,
475 including Eloquence. Go to court.
476
477 2. Have the ruling from Part One cited as legal prece-
478 dent.
479
480 3. Achieve the same outcome as Part One, demonstrating
481 that we have indeed set precedent that works against
482 Microsoft's legal department.
483
484 Implications of the outcomes
485
486 If we win both cases: Microsoft has the legal high
487 ground. Making a derivative of a copyrighted work us-
488 ing a machine-learning algorithm allows us to bypass
489 copyright licenses.
490
491 If we lose both cases: Microsoft does not have the le-
492 gal high ground. We have good judicial precedent
493 against Microsoft to use when filing suit for
494 Copilot's behavior.
495
496 Either way, it's an absolute win for free software.
497 Taking down Copilot protects copyleft from enabling
498 proprietary derivatives (and by extension, protects
499 software freedom). But if we accidentally win these
500 two low-stakes “test” cases, we still gain something
501 else: we can liberate huge swaths of proprietary soft-
502 ware, starting with speech synthesizers.
503
504 Update: on satire
505
506 This post isn't “satire through-and-through” like
507 something from The Onion. Rather, my intent was to
508 make some clear points, but extrapolate them to absur-
509 dity to highlight other problems. I don't think I was
510 clear enough when doing this. I'm sorry.
511
512 Copilot has been found to suggest significant amounts
513 of code that is dangerously similar to existing works.
514 It does this without disclosing obligations that come
515 with those works' licenses. Training a model on copy-
516 righted works may not be wrong in and of itself; how-
517 ever, using that model to generate new works that are
518 not sufficiently distinct from original works is where
519 things get problematic. Copilot's users could apply
520 proprietary licenses to the generated works, defeating
521 the point of copyleft.
522
523 When a tool almost exclusively encourages problematic
524 behavior, the makers of that tool should have put
525 thought into its implications. GitHub and OpenAI have
526 not demonstrated a sufficiently careful approach.
527
528 I don't think that “going after” a smaller player just
529 to manipulate our legal system is a good thing to do.
530 The fact that this idea seems plausible to some of my
531 readers shows how warped our perception of the judi-
532 cial system is. Even if it's accurate (I doubt it's
533 accurate, but I'm not certain), it's sad. Judicial
534 systems incentivise too much predatory behavior.
535
536 Corrections It's come to my attention that Eloquence
537 may or may not still belong to Nuance. Further re-
538 search is needed. Eloquent Technology was acquired
539 by SpeechWorks in 2000.
540
541 1 https://seirdy.one/posts/2022/07/01/experiment-copilot-legality/
542 gemini://seirdy.one/posts/2022/07/01/experiment-copilot-legality/index.gmi
543 2 https://openai.com/blog/openai-codex/
544
545 3 https://en.wikipedia.org/wiki/GPT-3
546 4 See Comment Regarding Request for Comments on Intellectual Property Protection
547 for Artificial Intelligence Innovation submitted by OpenAI to the USPTO.
548 https://www.uspto.gov/sites/default/files/documents/OpenAI_RFC-84-FR-58141.pdf
549
550 5 https://nitter.net/mitsuhiko/status/1410886329924194309
551 https://twitter.com/mitsuhiko/status/1410886329924194309
552 6 https://lucumr.pocoo.org/about/
553
554 7 https://github.com/id-Software/Quake-III-Arena/blob/master/code/game/q_math.c
555 At line 552
556 8 I doubt anybody worth their salt would count on a company to hold itself
557 accountable, but at least they tried.
558
559 9 https://github.blog/2021-06-30-github-copilot-research-recitation/
560 10 https://github.com/wunderalbert
561
562 11 https://github.com/espeak-ng/espeak-ng/
563 12 https://seirdy.one/a/eloquence.mp3
564
565 13 https://seirdy.one/posts/2020/11/23/website-best-practices/
566 14 https://en.wikipedia.org/wiki/Deep_learning_speech_synthesis
567
568 15 https://github.com/mozilla/TTS
569 16 https://doi.org/10.48550/arXiv.2112.02418
570
571 17 If the stars align, you could file an anticipatory suit against the company.
572 It's common for declaratory judgement regarding intellectual property rights.
573 https://en.wikipedia.org/wiki/Declaratory_judgment
574
575
576
577 Glenda adventure sirjofri
578 ____________________________________________________________
579
580 >> Glenda found herself in a dark forest.
581
582 Do operating systems dream of electric bunnies? Noth-
583 ing is certain about that, but it does not prevent you
584 to try to imagine.
585
586 Sir Jofri offers us a piece of fiction built out of
587 the reality of the plan 9 operating system. [1]
588
589 Where should this go next?
590
591 A story first published on the 9front Mailing List.
592
593 1 http://sirjofri.de/oat/tmp/glenda_adventure.txt
594
595
596
597 Space Weather Woman tgtimes
598 ____________________________________________________________
599
600 As she names herself, Tamitha Skov [1] is the Space
601 Weather Woman. You read it right! She have been do-
602 ing, since now close to ten years, forecasts about how
603 is space weather is going.
604
605 Just a nerd fantasy? Only a sci-fi artist on a peri-
606 odic one woman show? Not at all! Knowing what the
607 sun is blasting toward Earth can reveal more useful
608 than it looks. This includes:
609
610 o personnal safety for some plane flights at high lat-
611 titude.
612
613 o GPS communication, something happening in the pocket
614 of many individuals, some of them even unaware of
615 the involvement of satellites in the process.
616
617 o Long distance radio communication, which include Am-
618 ateur Radio operators, but also emergency services
619 and militaries.
620
621 o Something that Starlink did not invent [2] is
622 satellite-relayed communication, including satellite
623 internet and voice phone transmission. Actually a
624 lot of wind turbines are being given satellite in-
625 ternet, and see how a little disruption [3] in
626 satellite internet access can disrupt their opera-
627 tion.
628
629 And all of these fancy things are benefiting from Tam-
630 itha Skov's efforts as a researcher, but also by in-
631 forming in layman's terms what is going on outter
632 space.
633
634 >> Weather phenomena like coronal mass ejections, so-
635 lar flares, and solar particle events. [4]
636
637 Science is elegant.
638
639 1 https://www.spaceweatherwoman.com/
640 https://yewtu.be/c/TamithaSkov
641 2 WildBlue, Viasat, NordNet...
642 First amateur stellite launched in 1961.
643
644 3 https://hackaday.com/2022/06/02/the-great-euro-sat-hack-should-be-a-warning-to-us-all/
645 4 https://en.wikipedia.org/wiki/Tamitha_Skov
646
647
648
649
650 A C64 4chan Browser tgtimes
651 ____________________________________________________________
652
653 The sewers of Internet in a C64? The link appeared on
654 various IRC channels such as #electronics or #osdev,
655 and not one more word. The investigation is open.
656 [1]
657 1 <No_File> https://imgur.com/H36LTRV BACK 2 ROOTS!
658
659
660
661
662 I Hate Modern Technology ig0r
663 ____________________________________________________________
664
665 >> The "advance of technology" is a source of excite-
666 ment as well as frustration. ig0r gives us a crys-
667 tallised view of human stupidity offered daily by
668 technology.
669
670 Modern technology sucks. This might be me behaving
671 like a pathetic little angsty hipster or trying to
672 LARP thinking I'm somehow cool, but I think it's a
673 genuine problem.
674
675 Planned Obsolesence
676
677 Technology is being designed to fail.
678
679 Apple purposefully makes batteries fail on their de-
680 vices and solders them in such that replacing the bat-
681 tery on an older device makes no sense, forcing the
682 customer to buy a new device.
683
684
685 Lenovo's quality has gone down the shitter. Thinkpads
686 used to be thick, bulky, and rugged such that a cave-
687 man could use it in place of a club. New models bend
688 and creak, the hinges breaking after several years of
689 use while older models still run like new.
690
691 The reality is companies want people to consume tech-
692 nology, not use it. They care about making a profit
693 rather than giving users a good experience, hence poor
694 quality of manufacturing to speed up distribution,
695 consumption, and the filling of landfills.
696
697 Modern Software
698
699 Modern software is just bad. Here's a few reasons
700 why...
701
702 o It's idiot proof, in that I have little control over
703 settings and configuration
704
705 o Software has become synonymous with adware (see Mi-
706 crosoft putting ads into explorer)
707
708 o I have to pay money for it (fuck you, if I could
709 copy-paste a car I would)
710
711 Smartphones
712
713 Smartphones are the most annoying little shits, and
714 for some reason they've become ubiquitous.
715
716 Restaurants are starting to ditch regular menus in fa-
717 vor of QR codes to be scanned with smartphones. Why?
718 Paper is more reliable. This is a step backwards in
719 my opinion. What if I don't have a data plan? What
720 if I don't carry a smartphone?
721
722 Also why does everything have to be an app? Why does
723 my passport have to be an app? I'm perfectly happy
724 carrying around paper ID (paper ID doesn't spy on my).
725
726 People are idiots
727
728 Most companies justify making technology suck more by
729 saying it's 'easier' and more 'convenient' for normal
730 people.
731
732 Stop making easy and more convenient. Nobody asked
733 for that. We were happy when technology was hard.
734
735
736
737 Better recording of the IRC Now events ircnow
738 ____________________________________________________________
739
740 Here is a link with a better recording than the one in
741 the previous tgtimes opus [1]
742
743 As a teaser, here are some random contents from it:
744
745 o Independence from Silicon Valley
746
747 o Self-Governance with Free Software and Right to Code
748
749 o Live demo of OpenBSD system administration from the
750 ground up.
751
752 1 https://media.libreplanet.org/u/libreplanet/m/ircnow-of-the-users-by-the-users-for-the-users/
753
754
755
756 MNT Pocket Reform OS support tgtimes
757 ____________________________________________________________
758
759 All these laptop and portable devices come with either
760 Windows, Apple iOS or OSX, Android, sometimes Chrome
761 OS, and even more rarely Ubuntu installed upon.
762
763 But the open hardware commnity is rising, and calls
764 for a change. The MNT Pocket Reform lists more exotic
765 operating systems as officially supported, [1] or at
766 least acknoledged and listed in the front page:
767
768 o Debian GNU/Linux
769
770 o Support for other distributions: Arch, Ubuntu, Void
771
772 o Plan 9 (9front)
773
774 o Genode
775
776 o OpenBSD (in development)
777
778 Are we seeing a year of the open hardware laptop com-
779 ing?
780
781 1 https://mntre.com/media/reform_md/2022-06-20-introducing-mnt-pocket-reform.html
782
783
784
785 Darknet Diaries tgtimes
786 ____________________________________________________________
787
788 The mysterious Dark Net. While not an official insti-
789 tution, this hypotetical place built its very own
790 identity through popular culture and medias. Famous
791 and infamous, the depths of the limbos are explored in
792 the Darknet Diaries podcast, covering and reporting
793 the day-to-day events of that suspicious eden of sha-
794 dow. [1]
795
796 1 https://darknetdiaries.com/
797 https://en.wikipedia.org/wiki/Darknet_Diaries
798
799
800
801 The Modern Mechanical Turk tgtimes
802 ____________________________________________________________
803
804 In 1770, long before the exploitation of electricity,
805 a machine was built in the pretention of being able to
806 play Chess. This machine named Mechanical Turk was
807 nothing more than a moving puppet actuated by a small
808 human, such as a child. A child who is good at chess,
809 that is!
810
811 Actuating levers, the operator would make the puppet
812 move, fooling the audience that technical advances oc-
813 casionally make use of black magic.
814
815 Amazon called a software platform Amazon Mechanical
816 Turk. [1] It offers management for harvesting food
817 for machine learning: human description of images,
818 videos, products, and other kind of canned thoughts
819 that machine learning can make use of to build models.
820
821 Uber for Cyber. Human translators shouting at ma-
822 chines the language they got whispered through their
823 life.
824
825 Ghostworker. Noun. 1. Worker performing activity that
826 will only be appreciated as data feeding an algo-
827 rhithm. 2. Worker with no access to who it provide
828 work to, both employer and client are invisible to
829 him. [2]
830
831 given the very large scale at which these data-
832 harvesting structures are deployed, it means that you,
833 web user, have experienced the Google and Cloudflare
834 "captcha" block window. That window preventing you to
835 submit a form unless you click on all buses, track-
836 tors, crosswalks, traffic lights... to verify that you
837 are indeed a human and not a bot trying to access the
838 website. Instead of prooving its belonging to the
839 mankind, at the opposite, the user is explaining to
840 machines what is a bus, a tracktor, a crosswalk, or a
841 traffic light.
842
843 Here is your Great Technological Singularity for the
844 greatest common entertainment: Nothing more than a
845 moving puppet, actuated by humans, barely even paid
846 for it, if paid at all... [3]
847
848 1 https://en.wikipedia.org/wiki/Amazon_Mechanical_Turk
849 2 https://www.ghostwork.org/
850
851 3 https://en.wikipedia.org/wiki/Mechanical_Turk
852
853
854
855 Publishing in The Gopher Times you
856 ____________________________________________________________
857
858 Want your article published? Want to announce some-
859 thing to the Gopher world?
860
861 Directly related to Gopher or not, reach us on IRC
862 with an article in any format, we will handle the
863 rest.
864
865 ircs://irc.bitreich.org/#bitreich-en
866 gopher://bitreich.org/1/tgtimes/
867 git://bitreich.org/tgtimes/
868
869 Did you notice the new layout? We now can jump be-
870 tween single and double column as it is more fit: Some
871 large code chunks will not fit in a two-column layout,
872 but text is more pleasant to read on two columns.
873
874
875
876
Response: application/gopher-menu
Original URLgopher://bitreich.org/1/scm/tgtimes/file/opus6/tgtimes6.txt.gph
Content-Typeapplication/gopher-menu; charset=utf-8