Thursday, December 31, 2020

Re-orienting mid-journey (WP-->Blogger)

lego hiker with map and compass
Image by Andrew Martin at Pixabay

[Copying from My Wordpress to Blogger Google Doc]

2020-12-23

  • Preparing to move a quite long post I was distracted by 
    • Wondering and discovering auto TOC is possible - before remembering that just moving it all is the goal for now.
    • Contemplating blogger thumbnails & WP featured images - learned from MaryC.fromNZ and Deepak Kamat over at webmasters.stackexchange about ‘handling’ Blogger’s choice of image to use as thumbnail.  As Blogger would use the first of my own images in the post, either make sure that’s worthy of thumbnail or insert one higher up perhaps with some css style to hide it.

2020-12-24

  • Reflecting that already I glimpse trials and tribulations others’ mentioned. Already tweaking the plan, and fairly sure my creativity would not extend to entertaining anyone with my step-by-step, I postponed deciding whether to blog it, and started a Google Doc (basically turning the plan into a repeatable procedure, and capture reflections) 
  • Decided to prioritise posts most viewed at the old place, and finally worked out how to get a list from Jetpack of posts and their #views that I could copy over to the doc which conveniently links to each post. 
  • Third post copied (the post most viewed over there (4219 views! -- It did contain 65 external links! and no I started but didn't check them all) :
    • Minecraft in public libraries for teens and young adults @lgreenpd
      • pleased/surprised to find the kwouts are still deliverable, even though I can't find the developers anymore. 
      • Frustrated with Blogger editor's behaviour in toggling html/compose though: I could not switch to Compose without removing "</area>" ; yet everytime I had to switch to HTML the interface reinstalled "</area>" into the kwout map even though the post would not save because it claimed that was bad html!
      • Major complications copying over the Comments and Pings!

2020-12-25

  • Planning to move the fourth post, I started a Google sheet to note which I have copied and why and the difficulties each presented (+ to capture a bunch of numbers, of comments, pings, internal/external links, images, kwouts).
  • Copied over

2020-12-26

2020-12-27

  • Spent a bit of time thinking about the drafts never posted
    • Copy/pasted 3 page list of drafts to sheet, then cleaned the data, deleting columns, Find/replace extranei ; sort by Category, tag and then date because category and tag mean I once thought it relevant to include, recency is not so important.
  • Notepad refines the procedure = 
    • old in html, copy, paste into notepad, then cut piece by piece into new post in html

2020-12-31

Sunday, December 27, 2020

Image Use ethics - CC0 and photoscraping sites

Two issues 

  1. what is our responsibility for proving right to use images?
  2. can Google prevent photo scraping sites from landing higher in search results than original sources?

Should we say where we get an image even if it is in the public domain?

If we find an image via Google's image search with the licence Tool set to filter by Creative Commons licence?

I wanted an image to convey my mid-journey step in which I refined my plan - and I thought of orienteering with map and compass.  Found great lego orienteer images, but could not find who had made them available, (Google-search-filter)-allegedly by CC.  

The one/s I liked appeared (and I list here not really to shame, but to show how far I explored looking for the image creator):

With licence statement but no link or credit at:

  • (Copyright Creative Commons CC0) ecoparent  - so I quickly checked on what CC0 means, and learned that if one is not the creator one should not label it as this site did.

with no copyright/licence statement or photo claim/credit at:

Then I briefly thought I had found the source - where it appeared to have been released into the Public Domain with CC0...

Screensnip of the CC0 claim by scraping site, see it yourself at webarchive

 - and where EXIF data is given as if shared by the creator. But still I wondered - particularly as there is no id for the photographer even though I would have to register to download the image. Is the site (pxhere.com [ph]) legit?

So I googled that! 

 is pxhere legit

and found I was right to be suspicious: 

Photoscraping sites are not hard to detect.

Alan Levine shows how one scraping site had scraped an image from pixabay. A clue that a site is probably a scraper, he mentioned, is the lack of identification of the contributor.  Later in the post Alan discussed how a CC-BY image he had on Flickr had been picked up by [ph] who claimed it as CC0 - an outright lie.

Now I suspect [ph]'s short url indicates they've probably even renumbered photos they scrape so I couldn't just swap out the domain, so I picked a few keywords and searched pixabay. Sure enough, there the image appears to have been created and shared CC0 by Andrew Martin (aitoff) 

lego hiker with map and compass
Image by Andrew Martin at Pixabay

Aside - I am only 90% happy with the (optional for CC0) attribution phrase pixabay provides - they link fine to the user, but not at all to the image - so I edited the html, and hope they reconsider their default setting. 

Google lets scrapers float up

Although the first image-scrape Alan discussed were images given to the public domain, the problem he describes is that Google lets photoscraper sites float to the top of their search results.  I believe Google ought to find a way to prevent that.  

Regardless, I would prefer to see everyone cite where they obtain the images they use online - particularly organisational sites like schools! Can we consider it a moral responsibility? I do. I also do not believe the difficulty of finding the true creators of public domain images releases anyone of that responsibility.  

Creators: don't join scraping sites like [ph] 

"It is impossible to delete your photos and your account, support simply ignores all requests. As a photographer, I highly do not recommend this service, a lot of photos are used with copyright infringement!" -- Danila Perevoshchikov 3 March 2020 at Facebook

 

Friday, December 25, 2020

Creating a Circle Desktop shortcut icon

Short version (what I did):

  1. make circles with transparent backgrounds in Powerpoint.
  2. and layer one image over another in Irfanview
  3. then convert with IrfanView to make a desktop icon
  4. and only then learn about icoconverter.com which I know can do 3 and suspect can do 1

Long version:

Retrieve five year old (2015) posts about making desktop icons

.., find a tool I learned about then is apparently a deprecated project, so lots of relearning...

goal

I wanted to show and tell my "customised a desktop icon to my own Flickr" story but the dots are trademarked by Yahoo. I don't know if that would mean one couldn't show non-commercial images incorporating the dots - but I also don't want to spend the time checking, so I erred on the side of caution.

For illustration I contemplated other icon customisations. Say, place the ALIA starry thing in the centre of a Chrome circle for the shortcut to my ALIA PD Drive folder. But - mm no - Chrome is probs trademarked too. I might still do it to practice the technique, but won't show it :-(

I wanted to layer two side by side coloured circles over my scaredydragon and have the combination in a circle with transparent background as an .ico file.

What I did

So while I can't show you the end result, I can share videos to illustrate what I learned.  

Also note it took many rinse and repeats to select the precise portion of the picture (the dragon's face and ears) that I wanted to show in the eventual circle, into the necessary part of the intervening square. And for the memory file, that involved, in IrfanView:

  • load image (open folder) browse
  • (too small , resize up =Ctrl+R or Image > Resize/Resample
  • I only want his face, so Shift+C Create Custom Crop Selection size 256 x 256 *
  • --repeat those steps until the original resize was such that the custom crop grabbed what I wanted.
  • Saved that square (can't remember how)

* because:

"Microsoft recommends using an icon with a size of at least 256×256 pixels, to ensure it is properly displayed in the Large icons view"--Digital Citizen.

Next

  1. Paint's dropper tool let me grab the exact colour I wanted, to create two squares that I could copy paste into:
  2. Powerpoint where (thanks to Articulate 360 I now know how to convert them to circles with transparent backgrounds.

  3. While I couldn't work out how to save both together still with a transparent background in Powerpoint, Kevin Stratvert showed how to remove a background from a picture with Paint3D.

Back to IrfanView,

  • Edit Insert overlay/watermark image
  • ... centre,
  • 0% transparency gave me the effect I wanted

Next step back to Powerpoint to get another circle, but Powerpoint won't save it as an .ico.

So I went back to IrfanView to change the png to .ico, but the conversion seems to have left a jagged edge. Still ... I am super happy I have an ico that makes my desktop shortcut icon visually meaningful.

Update: I've since found the png-ico conversion is easier with ICOconvert recommended by Maraksot78 (video: How to Create Custom Desktop Icons (It's Easier than you Think)) (although as mine was just one image png to ico convert, as asked, I used the old ICOconvert homepage).  

When I look more at ICOconvert online it seems they also offer cropping and Styling which includes circles - so I wonder if the special shape styles come with transparency?  Maybe I'll try it out soon.  No hurry because I did try their tab for "Multiple PNGs to one ICO" but it had no upload mechanism.

Thursday, December 24, 2020

Bless Blogger embracing p tags

When Open Live Writer stopped working for me I sulked in hiatus. While submitting my latest CPD I felt the faintest spark of blogging urge again. And then came the reminder that I'll be paying for that other blog again soon. 

So I googled doubtfully for a Google-Blogger <p>shaped miracle. 

-site:pinterest.* "blogspot" paragraphs

{Aside: Query for Google: can their algorithms be clued in to prioritise pages dated after June 2020 on  any search including Blogger or blogspot?}

Guess the contours of my curiosity when I spied in the results, dated June this year:

Google's NEW Blogger Interface does not ... MAKING A MARK

@Katherine Tyrrell, Thank You. Not only has Katherine been so persistent with Blogger that she was around to experience the new interface - she also didn't hold back in her critique and she courteously linked to the Official Blogger Blog's (apparently tardy) Announcement. Although neither specifically mentioned p tags and appropriately spacing paragraphs, I cautiously began to hope.

Hope waxed and waned : 

Yay every ENTER a paragraph,
Oh No how do I get it to stop?,
meh Okay that's how to switch to html,
what's with the blank lines?,
oh great I can copy across the kwout,
OMG why does it keep re-inserting </area> inside the <map> every time I switch back to html-edit? Seriously, I have to go back to html to insert a <hr>? (Good grief!),
Hhmmm?
.....[eventually]... right well okay that wasn't a really bad nightmare,

So let's post how excited I am about the <p>-respect,

OMG why is it not now doing the paragraph thing?
Oh I have to switch that there from Normal to Paragraph...
every time? Why?
[and later] but not anymore?
WTF is going on?

And now hours have slipped away!