September 2017 - Talk - lists.gtalug.org

war story: mixing up UEFI and Legacy Booting
by D. Hugh Redelmeier 15 Sep '17

15 Sep '17

==== background (feel free to skip) "Legacy Booting" means the old fashioned way that PCs booted: - read the first sector of a disk into RAM ("the Master Boot Record" AKA MBR) - jump to that RAM copy: start executing 8086 instruction there. (This is the same way the original IBM PC operated and completely ignores memory management, 32-bit mode, and 64-bit mode. - the BIOS provides routines for crude I/O (but they only work in 8086 mode) - by convention, the MBR "understands" the traditional kind of partitioning, in part because the partition table is within the MBR itself (but only the four primary partitions). Thus it can chain to the boot record on the "active" partition. UEFI booting is more, uh, sophisticated. The UEFI firmware (it should not be called BIOS since it isn't a BIOS) understands GPT partitioning and it understands the FAT32 filesystem. There is a distinguished partition called the EFI System Partition that is FAT32. Or multiple such partitions, but that confuses at least Windows. The firmware can run programs from that partition in a fairly rich environment. Stupid regression: 32-bit UEFI can only run 32-bit programs; 64-bit UEFI can only run 64-bit programs. Why is that a regression? Because good old 16-bit BIOS never got in the way of 32-bit or 64-bit modes. But the 64-bit Linux now can run on a 32-bit UEFI system if the underlying processor supports 64-bit mode. Most modern PC processors can handle 64-bit mode. The exceptions seem to be dying (Intel's Edison comes to mind). For market-segmentation reasons there are a number of 32-bit UEFI systems at the low end but those have 64-bit processors (eg. low-end notebooks and tablets). Generally UEFI mode is better: + supported by all modern PC systems + naturally supports GPT partitioning and large drives + supports Secure Boot Why would you want to use Legacy mode? + you want to run MS-DOS or FreeDOS or BeOS or some other ancient system. Really: this is sometimes necessary if you want to flash new firmware and you cannot or don't want to use Windows for this. (But some systems allow flashing from the UEFI firmware itself.) + you don't want to learn about the new world of UEFI + you want to port an old installed system directly to new hardware (using old disks or at least old disk images). + you want to dual-boot with a system already installed in Legacy Mode. ==== foreground There is another use case for Legacy Mode: you accidentally install Linux in Legacy mode. This is especially unfortunate if you wish to dual boot with Windows that has been installed with UEFI booting. This is probably a big mistake, and I don't know a better fix than re-installing. This happened to my son. Why? >From a certain amount of experience, I've found that too many systems have firmware that is coy about exposing the UEFI / Legacy boot option. - some systems just plain don't offer Legacy Mode (eg. most Windows tablets). Problem solved. - on my son's Asus notebook, there is a firmware configuration choice "Launch CSM". That really means "Legacy Mode Booting". This confused me since I have several systems that have a similar choice that doesn't cause Legacy Mode Booting. - on my MSI Cubi's firmware the undocumented option is called something like "Windows 8/8.1" mode. - other systems I have allow more flexibility. The boot menu offers to boot a device either way if it looks as if it could be booted either way. - if you can enable or disable CSM, you must enable it for Legacy Booting but on most systems it doesn't force Legacy Mode booting. When Fedora installation medium is booted in Legacy Mode, it will install a Legacy Mode system; when booted in UEFI Mode, it will install a UEFI Mode system. It doesn't wave flags about this decision. It would be nice if Fedora noticed that it was installing a Legacy system on a disk with a EFI System partition and warned about this.

1 0

OT: IPv6 and Mac OS Lion (10.7)
by Ivan Avery Frey 14 Sep '17

14 Sep '17

Where can I get help troubleshooting network issues with Mac OS 10.7?

2 2

Imagemagick and exif data labels question?
by Russell 14 Sep '17

14 Sep '17

I have a relatively large number of photos which I need to print with the timestamp and also add comments on the face of the image. I'm using convert to run through the directory to apply the timestamp to the image itself. #!/bin/bash for file in *.jpg ; do convert "$file" -font <whatever> \ -pointsize 72 -fill white -annotate +100+100 \ %[exif:DateTimeOriginal] "new-${file}" done I'm also using convert to apply the comment by hand for each image. convert <image>.jpg -gravity South -pointsize 30 -annotate +0+120 'This is a comment.' temp1.jpg I would like to preserve my comments as exif data and pull them out to then label the images in batches using the first example as a template. Is it possible to define your own exif data and mung it into the image file? Or perhaps there is a gui tool which lets you view the image and also save a comment as exif for labeling? If I'm going to type comments I'd really like them to be part of the image record for future extraction at will. Also if someone has suggestions on how to style and make the labeling font stand out on colour images, which are not converted to greyscale, but are then printed in black and white, I'd appreciate them. Thanks. -- Russell Sent by K-9 Mail

3 3

"Massage Passing" library?
by William Park 13 Sep '17

13 Sep '17

Hi all, I'm looking for keywords to search for... I have various local peripherals that I need to read and write. From top of my head, I'm thinking 3 ways: 1. Use select(2) (and friends) to round-robin the peripherals. 2. Each peripheral is serviced by a separate thread, and main thread does the business logics. The peripherals don't need to talk to each other (but this may change). 3. Each peripheral is serviced by a separate process, and they pass "messages". This option is what I want to investigate. So, do you know any "message-passing" scheme, framework, or library that I can look up? I'm not talking about OS or kernel level. More at application level. -- William Park <opengeometry(a)yahoo.ca> .

6 5

Links from my talk on Web Scraping last night
by Myles Braithwaite 👾 13 Sep '17

13 Sep '17

Jupyter Notebook: <https://github.com/myles/2017-09-12-gtalug-web-scraping/blob/master/present…> Articles: * Remedies for Web Scraping by John Gregory, <http://www.slaw.ca/2013/10/21/remedies-for-web-scraping/> * Web Scraping and Crawling Are Perfectly Legal, Right? by Benoit Bernard, <https://benbernardblog.com/web-scraping-and-crawling-are-perfectly-legal-ri…> Python Libraries: * Selenium (good for download data from a JavaScript only website) - <http://www.seleniumhq.org/projects/webdriver/> * BeautifulSoup (HTML Parser) - <https://www.crummy.com/software/BeautifulSoup/> * Pandas (Data Analysis Library) - <http://pandas.pydata.org/> * Matplotlib (Fancy Graphs) - <https://matplotlib.org/> >From Questions: * Good Python library for easily sending emails - <https://github.com/lavr/python-emails> * Stewart's talk on OCR, A Bit More Than Mostly Searchable: Scanned Paper You Can Find - <https://www.youtube.com/watch?v=EH_txB_hJWw&t=9s>

1 0

Myles's Open Data Talk
by Stewart Russell 13 Sep '17

13 Sep '17

Great talk from Myles tonight. Canada is trying to be better about open data, but it has a long way to go. There are open PDF scraping tools, but they tend to be very limited in domain. It's always fun when you get data that must be published, but there's no stipulation that it's usable. The best example of this is the US Armed Forces Appropriations data that's published in a single giant PDF every year. While it genuinely does publish details of every military contract, it uses a fiendish set of page templates to make it very difficult to parse. I ended up making minor headway consider each page as geodata, with each page a map with words at given coordinates. Cheers Stewart

1 0

What time are we supposed to meet?
by Jon Stanley 12 Sep '17

12 Sep '17

Johnathan Stanley

1 0

Where are we eating?
by David Collier-Brown 12 Sep '17

12 Sep '17

I've not seen the traditional discussion yet... --dave -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest davecb(a)spamcop.net | -- Mark Twain

4 3

[GTALUG-Announce] Meeting on Tuesday at 7:30pm
by hi＠gtalug.org 10 Sep '17

10 Sep '17

<http://gtalug.org/meeting/2017-09/> # xterm and Docker *xterms: State of Art 2017* with Christopher Browne Many of us "live" in xterms, and there have been some developments. This talk pokes a bit at: - The old, and what's still being developed - The new, and there ARE some interesting developments - A bit of the crazy! Want an xterm that simulates shakey old amber monitors? You got it! *Web Scraping* with Myles Braithwaite TDA ## Location George Vari Engineering and Computing Centre 245 Church Street, Room 203 Ryerson University <http://goo.gl/maps/16oJ2> <http://www.openstreetmap.org/way/23447525> ## Schedule * 6:00 pm - Please discuss on the general mailing list (i.e. <talk(a)gtalug.org>) where you want to go for dinner. * 7:30 pm - Meeting and presentation. * 9:00 pm - After each meeting, a group of GTALUGers move to The Imperial Pub (54 Dundas St East) for refreshments and more socialising. # Code of Conduct We want a productive happy community that can welcome new ideas, improve every process every year, and foster collaboration between individuals with differing needs, interests and skills. We gain strength from diversity, and actively seek participation from those who enhance it. This code of conduct exists to ensure that diverse groups collaborate to mutual advantage and enjoyment. We will challenge prejudice that could jeopardise the participation of any person in the community. The Code of Conduct governs how we behave in public or in private whenever the Linux community will be judged by our actions. We expect it to be honoured by everyone who represents the community officially or informally, claims affiliation or participates directly. It applies to activities online or offline. We invite anybody to participate. Our community is open. Please read more about the GTALUG Code of Conduct here: <http://gtalug.org/about/code-of-conduct/>. If you have any questions, comments, or concerns about the GTALUG Code of Conduct please contact the GTALUG Board @ <board(a)gtalug.org>. --- GTALUG Announce mailing list announce(a)gtalug.org https://gtalug.org/mailman/listinfo/announce

1 0

Voynich Manuscript update
by Richard Weait 10 Sep '17

10 Sep '17

Now, a claim that the Voynich Manuscript has been "decoded". Well, "read" really, not decoded. It was in shorthand, of a sort. https://arstechnica.com/science/2017/09/the-mysterious-voynich-manuscript-h… William Porquet did a talk on the subject at TLUG a few years back.

2 1