Google News alternatives for 2022

Google News is becoming ever-more picky about what it will display in search. A special problem is its growing lack of full-coverage in terms of timeliness (i.e. it defaults to old news after just a few timely items, when there should be many more present). A clear example: apparently no outlet in the entire world has once mentioned Staffordshire -police -dog in the past three days. Yet Bing News has the ‘missing’ news stories, and from the same search… so why doesn’t Google News?

Another major problem is its seeming inability to weed out all the ‘market reports’ pseudo-news spam. There are so many of them because I presume they’re an entry-point for scams involving sales of worthless stocks and shares. If you’re gullible enough to buy one of these reports, then you set yourself up as a ‘mark’ to be conned.

What then are the search-box alternatives to Google News at the end of 2021? There are not many, but here they are.

The best:

Bing News. This is by far the best bit of Bing. Also the best news search service for timeliness, with ‘sort by date’, and the ability to pick up and display news even from the last few minutes. Also has very good coverage of local and regional news here in the UK, able to pick up the Bogglinton-on-the-Wold Bugle class of local news-sheets and more. Though they have recently started including ‘local classified ads’ as news, which is slightly annoying. The search box accepts two -knockouts, e.g. comics digital -nft -nfts and respects these in results. A third -knockout keyword will be ignored.

Proquest. UK public library members can get free home access to this, and can get full text from it (though no pictures, which some may consider a pleasing feature). Search major newspapers and selected magazines by multiple keywords, with many facets. Speed is reasonable for something that’s not a proper search-engine. Can re-sort results by date and narrow by date using a simple slider. Does not respect -knockout words, but does appear to accept NOT as an operator. Very little timeliness, but good for trawling the last year, especially in the UK and the major corporate U.S. newspapers and magazines. In the UK it gets down to the Bogglinton-on-the-Wold Bugle level.

Knewz. A self-proclaimed “Google News competitor” from heavyweight publisher Rupert Murdoch. A few sources reported it was “killed off” in summer 2021 and Wikipedia now claims it’s dead and gone. But it obviously isn’t dead, and Knewz is still working fine for me. I had no idea it even existed until now, so the first PR team on it obviously did a very bad job. It can’t re-sort results by date, but a timeliness indication is given on results. Has a definite bias toward US sources, with the main UK corporate newspapers and BBC News seemingly the only British sources. Supposedly favours conservative sources, but curiously it can’t find last month’s headline stories from The Spectator (UK) or Reason (USA). Possibly it considers them magazines rather than news, yet it does index magazines like Esquire and Wallpaper. Has a horrible yellow colour (which I guess might be fixed with a CSS tweak), but at least there are no blaring photos on results…

NewsNow. A useful ‘at a glance’ timeline for your search terms or phrases, and can surface things that might have gone unnoticed… e.g. “Wiley buys Open Access firm Knowledge Unlatched” — who knew? Only the British book-trade journal The Bookseller and the Library Journal, it seems. Sadly NewsNow does not respect -knockout keywords, and instead adds them. Pop-ups and overlays. A search for digital comics gives only two results, confirming the feeling that the range of sources is actually rather limited. Still, the trade magazines and original press-releases it indexes may make it useful to some.

The rest:

DuckDuckGo News. You get to this via a standard search and then switch to the “News” tab. Basically, it’s identical to Bing News, though with what seems to be less sources and less timeliness. So it’s better just to use Bing News instead.

Yahoo News. Has several major problems. No ‘sort by date’. Does not respect my -knockout keywords. e.g. comics digital -nft -nfts gets a wall of stories from hucksters touting comics and NFTs, and artists warning against buying NFTs of stolen art. Still, these results are at least relatively timely.

Yandex News. Difficult to control for timeliness. Search for just one keyword, and you’re fine in terms of timeliness… but results are incredibly scattergun. Search for two keywords and you get nothing at all from 2021 and lots of way-back items. Search for a phrase in quotes (e.g. “digital comics”) and you get one story from 2019 and lots from 2010. Bizarre. Also, you need to translate.

Baidu News. Slow. Does not respect -knockout keywords (see the same problem with Yahoo, above). Appears to cover only approved Chinese sources. You need to translate.

Gab Trends. Has a strange political bias, supposedly to the right, which appears to exclude genuine conservative magazine titles and yet includes Russian propaganda outlet RT and fruitcake sites. Still, it offers search and if you know what to ignore you might get some mileage out of it, as it’s obviously capable of timely coverage.

More UnSplash

A few years back I made three curated picks from the Unsplash CC0 Creative Commons collection, and posted these here…

* Libraries and archives theme.

* Creative Industries I theme.

* Creative Industries II theme.

The CC0 status was later changed. For instance, Alex Knight uploaded his Robot picture under CC0 Creative Commons, prior to the end of 2017. But Unsplash now has its own licence, under which it is claimed this formerly CC0 Creative Commons picture now sits.

The new licence is not that bad actually, on glancing through it… it seems to just prevents the big stock companies from ingesting en masse and re-selling. But now comes the news that the evil stock agency Getty Images is about to “acquire” UnSplash outright anyway.

So here’s another pick, and under the still relatively permissive non-Getty licence, before the purchase goes through and any changes start happening.

What follows is ‘Creative Industries III’. As before, images reduced a little to make them more wieldy, and a few spamming brands (Apple, Nike etc) were airbrushed away. Photographer names are in the file-name, and should be credited if used in print etc. A few of the older pictures (see my collections above) don’t pop up for me in search, where you might have expected to see them again, and may have since been deleted or moved. Useable “draw-on-the-screen” images are very rare, but I found two good ones.

---

Creative Industries III:

Daylight code writing (useful, as most such pictures are on black):

All-night code writer:

Guitar maker:

Leather crafts maker:

Glass crafts maker:

Pattern knitter:

Children’s book illustrator:

Graphic illustrator:

Special-interest magazine editor:

General designer using a Cintiq or XP-Pen:

Table-top RPG game designer / miniatures painter:

Pinball table designer:

Children’s party clowns:

Branch librarian / local documentary-maker:

Field researcher (a bit spammy, re: the brand, but the closest I could get which says ‘field research’):

Local history writer:

Philosophers / old books conservator:

Sports vehicle designer:

Vehicle livery colourist:

Hair stylist:

Local TV studio, junior camera operator:

Local TV broadcast station desk-jockey:

Live theatrical event desk-jockey:

Podcast interviewer/presenter:

Synth/trance musician / YouTube celeb:

Young stage drummer:

Young comics reader:

Children’s creative dress-up:

Community dance:

Censorship:

JURN fixed and repaired

JURN is now as up-to-date as it can be, ready for the “back to university” crowd. I’ve completed a link-check of the full URL base, checking for continued presence of an indexed URL path in Google Search. The full path is checked, not just the top domain (e.g. foobar.foo/foo-foo/journal_of_foo/articles/ and not just foobar.foo). This checking process has been slow, taking about 18 months, on and off.

Of course, a few URLs may still have newly broken in the meanwhile. But the core URL base is kept fresh by a regular check of the key home-page URLs, as organised and listed at the JURN Directory of arts & humanities journals (English-language journals only). This Directory was link-checked and updated in mid September 2020. Also recently link-checked, back in July 2020, was JURN’s openEco Directory of over 800 journal titles variously related to the study of wildlife, ecology etc. Please update any local copies you may be keeping.

It’s not Funai

It appears that someone is using Nigeria’s Federal University Research Archive (FURA) to run one of those “produce a poisoned spam-PDF for every book-title in the world” robo-scams, and is using the repository server to spam these online with a high search ranking. See the results via a Google Search for…

site:dspace.funai.edu.ng horror

For now I’m keeping dspace.funai.edu.ng in my GRAFT (‘search across all the world’s academic repositories’), with the suggestion that GRAFT users simply add -dspace.funai.edu.ng to the end of their search to clear this spam. Hopefully the problem will be sorted out, and the server cleaned up, relatively soon. The server is not in the JURN index.

Duck flies

This is interesting. For once, DuckDuckGo’s “last week” search is now better than Google Search’s “last week” search. I have the same URL blocklist running on both, so there’s no difference there. But having run a regular search on Google, I found myself going through the exact same with the Duck and saying… “why did Google Search not get that, and that, and that…” And most of that was currently-dated blog or magazine posts, so there was no sense of items being dredged from the past and presented as if new. Of course, it could be a simple mis-match in terms of timeliness, and I just caught Google in a late-August “techies have gone to the beach” slacker mode. But I suspect not.

The borrowed page

I see that the Archive.org now has a ‘no-Borrow preview page’ on search-landing pages for its ‘Borrow’ books. They’ve also announced that their extended “Borrow” feature is to come to an end, as was always to be the case after the emergency period had passed.

One wonders if this new ‘page preview’, similar in nature to Google Books, is about bringing Google in on the new lawsuit from publishers? ‘If this goes down, so does Google Books?’ Just my guess.

On auto-downloading open access books

Martin Paul Eve has a new post on Zotero and auto-downloading open access books…

all I really wanted was to be able to embed an ISBN and a citation_pdf_url and have Zotero do the lookup and save the file. However, out of the box there is no easy way to do this.

His test book is quite interesting, his own new Close Reading with Computers: Textual Scholarship, Computational Formalism, and David Mitchell’s Cloud Atlas (April 2020), which applies textual computing to the science-fiction-philosophy novel Cloud Atlas.

I don’t know about or use the current version of Zotero, so I’m unsure what advantages it confers. I assume Eve intended to find a way to automatically harvest all CC-SA books in PDF, and build a local collection for automated analysis.

But I see his book is already on the OA book aggregator catalogue OAPEN. Theoretically then, since OAPEN is comprehensive and timely, one could have a harvester look at all the pages hanging off library.oapen.org/handle/ and save out only those pages with the required permissive CC “Rights” label on them. These pages each have a uniform PDF link URL in their HTML, in the form of library.oapen.org/bitstream/ and these could be easily extracted to a list. One would end up with a set of PDF links for a linkbot, ready to download to a local folder for computational analysis. I presume that’s what Eve intended to have Zotero do.

One would need to reference the OAPEN record page first, in the way I’ve suggested, since the PDF itself can have different or non-uniform or contradictory licence information. For instance in its interior Eve’s book is labelled as both “©” … “No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.” and also “Creative Commons Attribution-ShareAlike 4.0”.

How many items on OAPEN have a creativecommons.org/licenses/by-sa/ “Rights” label at present, as Martin’s book does? A Google site: search suggests around 650 titles. Half an hour of my filtering the OAPEN CSV suggests it’s actually just over 3,000 under some form of permissive CC that permits commercial use. That’s still a manageable harvest at present. But as the supply of OA books and monographs grows rapidly, the likely result of various OA mandates in the near-future, it might be a useful time-saver for text-miners and digital humanists if OAPEN were to maintain a single torrent of all the PDFs. Inside which a half dozen folders would neatly organise the books by CC licence type. Such a one-click solution might save a lot of faffing around with digging into and filtering their XML and CSV feeds, wrangling with harvester scripts and timeouts, or trying to wrestle with third-party services such as Zotero. A torrent could also save OAPEN’s bandwidth.

Dead Carrot

The Carrot2 search-engine appears to be kaput. It’s been down for days now, unable to show results. Multiple different browers, on different systems, even different VPNs, give no results. A query can be input, but… no results. Even a TOR browser can’t get it to respond. Such a pity, it was so promising.

The worthless Peter Lowe ad-blocking list

The Peter Lowe ad-blocking list is obviously now worthless, due to its over-reach and scattergun blocking of all sorts of legitimate things. Back in June I found it blocking Harvard. I’ve since found all sorts of similar blocks on things that should not be blocked. I’ve unsubscribed my browser’s ad-blocker addon from the Peter Lowe list, and I suggest that you also consider doing so.

Carrot2 search – a new script for multi-column results

The Carrot2 search-engine has changed both URL and layout. It was at search.carrot2.org/stable/search and it’s now just at the search.carrot2.org URL. I guess the public-facing search may have come out of beta? This is what their new layout looks like…

Not good, on a widescreen desktop monitor.

Which means I’ve made another custom CSS for it. This slices the Carrot into a multi-column layout suited to a widescreen 1920px monitor. It works in the Stylus browser addon, and you need to tell the script to target the search.carrot2.org site

Yum, crunchy Carrot! All the z-depths are set up nicely, so you can still click on/in the filters and search box.

I like to read the URLs in search-results and so I’ve turned them dark green and wrapped them to make this possible. The results look good with URLs that line-wrap by up to three lines. But it’s unavoidable that some very long URLs will wrap over four lines, and will thus spill over the element below.

In most cases there should be no need for any scrolling.

It only works in Day mode, with the Carrot’s new toggle-able Night and Day mode…

If you want a night mode, you’ll have to manually change the colour chips to a charcoal black on…

body, body.light {

and

div.ResultList

The other drawback is that whereas the old multi-column fix showed about 25-30 results, now we’re down to about 16. If you want it up to 20+ add the following code…


.ResultClusters {
display: none;
}


There’s also the lack of a “more…” button, but this usefully forces the user to use the Carrot’s innovative faceting systems over in the left-hand pane.

---

To install my fix, simply go to Carrot2, then left-click on your icon for the Stylus browser addon and click “Write style for…”.

Then paste in this…


/* ==== CARROT2 - Multi-Columns v.02 Oct 2019 ==== */
/* run this Stylus script on search.carrot2.org */
div.document div.url {
overflow: hidden;
color: #3a7730;
font-size: 110%;
}
body, body.light {
background-color: #ece5db;
}
.ResultList > div > a > span.url {
color: #3f7126;
font-size: 80%;
font-weight: bold;
white-space: pre; /* CSS 2.0 */
white-space: pre-wrap; /* CSS 2.1 */
white-space: pre-line; /* CSS 3.0 */
white-space: -pre-wrap; /* Opera 4-6 */
white-space: -o-pre-wrap; /* Opera 7 */
white-space: -moz-pre-wrap; /* Mozilla */
white-space: -hp-pre-wrap; /* HP Printers */
word-wrap: break-word; /* IE 5+ */
}
a {
font-size: 100%;
}
.SearchForm {
z-index:20;
}
div.ResultList {
z-index:4;
position:absolute;
background-color: #ece5db;
padding-top: 180px;
top:0;
bottom:0;
right:0;
column-count: 5;
width: 70%
}
.ResultList > div > a >div {
font-size: 80%;
}
div.sources {
padding-top: 25px;
}
div.ClusterList {
background-color: #ece5db;
padding-top: 10px;
column-count: 2;
}
div.clusters-tabs {
width: 40%
}
div.clusters {
z-index:4;
width: 50%
}


Tested in Opera, which is a browser that runs on Chrome. It will probably work with other CSS style injectors.

Be warned that Carrot2 will perma-block an entire IP address (in the case of BT in the UK, or VPNs, that can mean hundreds of thousands of users) if it detects “excessive traffic”.

---

UPDATED (November 2021):


/* ==== CARROT2 - Multi-Columns v.03 Nov 2021 ==== */
/* run this Stylus script on search.carrot2.org */
div.document div.url {
overflow: hidden;
color: #3a7730;
font-size: 110%;
}
body, body.light {
background-color: #ece5db;
}
.ResultList > div > a > span.url {
color: #3f7126;
font-size: 80%;
font-weight: bold;
white-space: pre; /* CSS 2.0 */
white-space: pre-wrap; /* CSS 2.1 */
white-space: pre-line; /* CSS 3.0 */
white-space: -pre-wrap; /* Opera 4-6 */
white-space: -o-pre-wrap; /* Opera 7 */
white-space: -moz-pre-wrap; /* Mozilla */
white-space: -hp-pre-wrap; /* HP Printers */
word-wrap: break-word; /* IE 5+ */
}
a {
font-size: 100%;
}
.SearchForm {
z-index:20;
}
.ResultsScreen {
height: 100vh;
display: grid;
grid-template-columns: auto 0.3fr .75fr;
}
div.ResultList {
z-index:7;
position:absolute;
background-color: #ece5db;
padding-top: 10px;
top:0;
bottom:0;
right:0;
left:0px;
column-count: 4;
width: 100%
}
.ResultList > div > a > div {
font-size: 80%;
}
div.sources {
padding-top: 5px;
}
div.ClusterList {
background-color: #ece5db;
padding-top: 10px;
column-count: 2;
width: 100%
}
div.clusters-tabs {
width: 40%
}
div.clusters {
z-index:4;
width: 100%
}