
BWA Hahahahahahahhahhhahhahahahha
*Gasp*
Hahahahahahahahahhahaha...
*Wheeze*
Hahahahahahaha...
*Wipes away tears of incredulity*
Bahahahahahaha!
*Dies laughing*
Microsoft's Edge browser is the subject of an amusing new bug report, alleging it somehow manages to screw up printing strings of numbers. The report on Microsoft's developer portal describes the issue where PDF files printed through Edge will display numbers and text incorrectly when exported. "Edge displays PDF correctly …
"What do you expect from a company that went from Windows 8 to 10?"
Of course! They were using an early version of Edge while developing it!
(See also: Microsoft's claims about how successful Windows 10 deployment has been, while world+dog was avoiding it at all costs.)
"who might have reason to send an image to PDF."
if you buy something online, and they show a receipt in web format, sometimes it's best to make a copy, and PDF makes more sense than printing. I can't imagine a WORSE outcome than having the WRONG numerical information on that receipt when you try and track your package that didn't arrive for some reason...
"I can't imagine a WORSE outcome than having the WRONG numerical information on that receipt when you try and track your package that didn't arrive for some reason...
How about printing PDF boarding cards for airline flights? Bonus points if you have a foreign name or the wrong amount of melanin.
Like in Firefox, with the PDF tick widget added to Print Prevew by the "Print Edit" plugin; this creates mixed text and image PDFs, unlike the retarded, rendered-image-only PDFs of all PDF virtual printer drivers, which prevent later editing, text extraction, link use, and rendering re-flow!
Print Edit is wonderful for deleting the surrounding banners, adverts and social media strips, and other bloat from pages before printing, although some idiot web designers included F'd Up divs which prevent re-flow across several print pages, so truncate printing!
@Sampler
Until I read your comment I'd just thought, "Who cares" and "Why would you even want to".
But you've hit nail on head here.
Because the new Windows S**t version will tie users ( read "victims") to whatever Microsoft allows, making things like using Edge to create a PDF more likely.
Currently there are probably more ways to create a PDF of anything than there are ways to create the original content.
Publisher is a nice toy, but still a toy. It lacks several features to be a real DTP software, and also many have been removed lately. Because now most software is aimed at the millennials generation who are "digital natives", up to the point that many advanced features have to be removed form software otherwise they got lost and scratch their heads if they see things like CMYK and ICC. You take a selfie and publish it on Instagram with a cool filter, what else do you ask for???
Yet, I don't know how many Publisher users have color-managed workflow, including profiling monitors correctly. AFAIK PDF/X allows to embed (color calibrated) RGB images into a document with a CMYK output intent (yet, can Publisher crate them?), but then the printer RIP needs to support such features.
"now most software is aimed at the millennials generation who are "digital natives", up to the point that many advanced features have to be removed form software otherwise they got lost and scratch their heads "
Indeed. "Digital native" seems to be a euphamism for clueless idiot who knows how to use simple social media apps but doesn't have the first clue about how any of the tech works underneath. Analogy: They can drive the car (if its an automatic) but they have no idea how to check the oil never mind change it.
Personally I think the current generation despite being associated more with tech are a bunch of ignoramuses as far as its concerned. Hardly any of them (going by interviews we've done) seem to know any professional programming language such as C/C++/C# or java in any depth (as for assembler, yeah, right!). If we're lucky they might know python but normally their idea of coding is HTML,CSS and some junkscript to create a pretty web page. FFS.
The problem is that RGB deals with light, so 100%, 100%, 100% is white, while CMYK printing deals with ink, and 100%, 100%, 100%, 100% is black. CMYK 100%, 0%, 0%, 0% is also black, so is 100%, 10%, 10%, 0%, and 0%, 100%, 100%, 100% is pretty close to black too.
The gamuts (range of colours) are also completely different, CMYK is much more restrictive than RGB. A particular problem is bright light green, which is easy with RGB but impossible with CMYK.
To convert from RGB to CMYK you need an algorithm that works out how much ink is needed of each standard colour to make it look on paper like the RGB coloured light on a screen. There are many different algorithms, and each algorithm has many parameters you can set to compensate for paper type, etc. The only way to have accurate control over printed colours is therefore to work in the CMYK colour space. RGB is too vague.
Then you get "spot colours" where an additional ink is added where the ink itself is a custom colour...
Converting pure white and pure black is easy enough, the problem is everything between. First, a printer will charge you if it has to process your files into the required format. Second, you will usually have very little control on how your files will be processed.
Professional software allow you to soft (and hard) proof the results. The destination color space is emulated on screen (or prints from your local printer) to show the actual results (it's still an approximation, but usually good enough to avoid a lengthy and expensive process of fixing issues from actual prints). Just, remember you need profiles for the inks and paper that will be used.
Professional printers usually use standard ones (i.e. SWOP, FOGRA, etc.), and you would like to prepare output for them.
It's funny that while Microsoft is adding better support for color management in Windows, and better monitors are becoming available at lower prices, these features are being removed from Microsoft software.
It really looks Windows and its applications are being made dumber and dumber as if only some sort of "consumer zombie" is designed to use them. Once I would have found Publisher a welcome addition to the Office suite, now it's mostly a useless toy.
Hope Serif will release soon the Affinity version of PagePlus, to fill the niche of low price DTP software. InDesign CC is too expensive for occasional use.
Publisher kind of ended up as the poor relation to Word, though. Word moved from being a pure word processor into a kind of bastard hybrid between WP and DTP software around 2000-2005, leading to it being severely over-complicated for the former and still not being capable of the latter. And Publisher just kind of rotted on the shelf as Word absorbed it's purpose in life.
The end result is that MS now have no serious DTP package and 2 laughable ones.
Also, a good few years back MS decided that for home users Publisher wasn't to be included in the Office bundle. Why, God alone knows. I'm not sure Microsoft did. I equate this with the point where they stopped taking any notice of what customers would want. But for SoHo users not having Publisher meant using Word instead. And also there was a host of budget DTP offerings that could be used to create a church or club flier, or a quick poster, or a birthday card and so on. And of course the home users who then found themselves needing to knock out a bit of quick DTP when they were at work didn't go to Publisher, if they were used to using Word. I wouldn't be surprised if many of them even got their employers to buy Serif's offering even though they had Publisher already.
"The end result is that MS now have no serious DTP package and 2 laughable ones".
Looks in file cabinet for CD's of Office 2000, XP and the last really good version of Office, Office 2003. Fondly remembers no ribbon. Wonders if a VM Ware Virtual Machine of 7 with Office 2003 would run faster than a machine with 10 Supserspy Edition and the latest Office 2016? I suspect Office 2003 will not install on 10.
I haven't yet tried to install Office 2003 in Win 10, mainly because the last time I tried to install Office 2003 on a computer, Microsoft wouldn't allow activation. I did run across a copy of Office Professional for Windows 95 - that jewel of software drove me batty a couple decades ago because of a "feature" that would reindex all documents edited once a day. The only problem is some users had edited files on a strange thing called "floppy disks". If a floppy was not left in the drive at the time the reindexing occurred, the computer would freeze and would only respond to a hard boot - which was easy then because most computers then had a real power switch and not a power request button. I still don't trust any Microsoft product that requires indexing or maintaining a database. I've had too many problems with products from Microsoft that attempt to manipulate databases. Now it appears their operating systems require some type of indexing to even function, but I could be wrong on that.
"Once I would have found Publisher a welcome addition to the Office suite, now it's mostly a useless toy....InDesign CC is too expensive for occasional use."
Back in the mid-90s after I finished a group training of MS Publisher, one of the clients, a young girl, burst into tears. When I asked her what the problem was she explained she had been tasked with producing a monthly newsletter and that she now knew she'd never have enough time.
So, we did a cost-benefit analysis of using MS Publisher versus Pagemaker. A Pagemaker licence back then was well north of $AU1,000 and hiring me for a one-on-one training a further ~$AU400. Publisher may have been "free", but time is money. The cost difference looked like being amortised in 3–6 months.
As with most of my Pagemaker clients, the YL brought her first finished work for my perusal and very good it was, too. It helped that we had created a number of templates for the job during her training. The time-saving exceeded our original estimate.
The first book I created with InDesign more than justified the cost of the licence. The time-saving versus a low-end DTP tool can be quite dramatic. The book had over a thousand footnotes, only one of which had the required full stop at the end. Putting those in manually would have been a chore, but the GREP in InDesign made that a trivial task.
Hint: I purchased my first Pagemaker licence second hand for 20% of the RRP. That meant I paid very little more than for a "new" low end product.
"It really looks Windows and its applications are being made dumber and dumber as if only some sort of "consumer zombie" is designed to use them."
But, hasn't that been the case with all consumer digital technology over the past two decades?
I will admit, 132 is also very rude, but by the time you have managed to count that high, you have forgotten why you were doing it.
Binary doesnt seem to be taught much in schools these days (at least around here), so using it really confuses some people; even "Maths" teachers have been caught out by the question:-
"How high can you count just using your fingers and thumbs."
Most answer "10", and have to be nudged to understand the correct answer is 1023.
... that it had a device independent API for graphics. So you could write code which displays something on your screen, and use the same code to write something to your printer, without having to care about what kind of printer it was or even if it was a printer or a plotter. The operating system would do its best to give you the same experience on any device...
... but that was in the 1990s and WinAPI (now known as Win16) now seems to be depreciated.
It would have been even better if the API featured all the bloody functions that one needed when it came to printing and rendering. Hell, if printing used the same units throughout it would have been an improvement what with having to pass much code through various twips to inches to cm to pixel translators before using them.
However even aside from this, where MS really dropped the ball is that they could have easily implemented an OS level printer preview that would have worked for applications using the Windows printing mechanisms, as distinct from using PostScript direct. Unfortunately printing has always been the poor cousin of PC software design and even now printing support both at the Operating System and application level is pathetic. When was the last time you didn't scream at your web browser's half-baked and largely useless print of a web page? Have you tried printing an image using the Windows 10 native image viewer (hint: it doesn't work, right click file and click Print to print usng the old image viewer's print handler which does work, most of the time).
"Eh, I think that you'll find that Display Postscript was developed by Adobe and NeXT. Before Quartz."
Display Postscript was once upon a time used by VMS, but disappeared from the product when the licence with Adobe expired.
VMS had lost the battle for the workstation market by then, but Display Postscript was missed by those of us who still found it useful.
Nutty idea anyway. I disable that on any browser, save PDF and use a REAL PDF viewer (that's secure and won't access internet, or cause browser to freeze due to enormous size).
I don't use Edge, but really this bug wouldn't bother me.
Windows: Foxit (Or Ghostview to extract images)
Linux: Xreader
Some years ago there was a problem reported with some Xerox printers that were printing some small font numerals as different numerals. IIRC the software was actually digitising the scan and reading it (partly to ensure banknotes weren't being copied) and then rewriting it to the actual print engine, and if the numbers were small enough they could be read incorrectly.
This therefore makes me wonder if Edge is reading the rasterised image sent to the screen and translating that for printing and not the source document. Does anybody with a better understanding have any ideas?
It sounds like the Xerox issue was that the scanning mode used a lossy compression algorithm, and that algorithm had trouble with digits. You scan in a "8" followed by a "6", the algorithm decides that, effectively, both are "8". Not the kind of space saving the user expects!
...and Lord save me from having to stitch together screen caps of another browser window in Photoshop.
Firstly, +1 on the just download a pdf and open it outside the browser. I'd give this as general advice more often if I didn't catch people downloading a new copy every time they want to open it up.
Second, PDFs just need to go away. The technical details of the format are HORRIFYING. If you gave me a copy of the spec, an open code reference implementation, and 20 consulting programmers from the original Adobe team I still couldn't make a version that was compatable, accurate, and secure. It would only be LESS horrible. Just. start. over.
Lastly, on browser based printing in general, why is this still in the perennial dark age of computing?
Mozilla had an unfilled position for the maintainer of the browsers print functions for years. Safari has a reader view, but still can't handle page breaks or let you adjust the layout to print properly. Then there is Chrome. Even though it hijacks the whole print system, it still can't format a page properly, or handle adjusting the printer settings. Add the unnecessary round trip up to Googles servers and back as a bonus.
This is neither an accident or incompetence. It is malice.
Microsoft owned up to further bugs in its leading and much loved Windows 10 product, whereby, on certain dialogue boxes, clicking 'No', 'Cancel', 'Back' or 'Close' was interpreted as 'Yes, Sure, Take What You Need And Do What You Will!'
When pressed, a Microsoft spokesperson issued the statement, 'Yeah, whatever...'
Microsoft is not alone here: a similar bug, but without the accompanying UI, surfaced in systemd.
..we automatically switch the default browser to IE from Edge. While I usually use a simple script to set the defaults for web browsing to IE and viewing PDFs to Acrobat Reader, sometimes I have to switch manually and I derive a perverse pleasure in ignoring the "Check it out" message from Windows and clicking "Switch anyway", not giving Edge a chance.
.. and that's the idea that there is something actively reviewing and possibly rewriting data sent to the printer (probably worth checking if it does it during "save to file" as well to address any attempts to preserve a webpage). We only know it now because it went haywire, but WTF is something like that doing in the code?
Changing election results? Making it look like posts negative over Microsoft get more downvotes than in reality?
Why does a browser need a function that manipulates its output?
99% of Windows users live in China.
99% of these buy their copy on the street for 10 Yuan, 15 if you add photoshop.
100% of these copies are the Ultimate versions - why waste good money on low class shite?
The PRC government has recognized this by issuing all official online documents and fill-out-forms as .doc only, so nobody even knows what .pdf is, anyway.
Note: Since Linux is free, why even use it? To save 98.87 (or whatever) USD is much more fun!
Soooo, this is a bug that you have to go hunting for on different systems?
"Edge displays PDF correctly but printed content differs notably," the bug notice reads. "Printed content depends on selected printer, on printer settings, and on used computer (please try a different setup if first result looks correct)."
So, you could try 1000 systems and not have this issue?
MS, (as other devs do), try to cater to as many different combinations of systems/software/printers etc., and miss a couple... So that explains all the vitriol here for MS?
I use whatever works, whatever OS etc., that gets the job done...
So, there is a bug that happens on *some*, *probably very few* computers... Welcome to the world of PCs.
I think a lot of people need to pull their heads outta their arses and understand that *nothing* works perfectly, no matter how loved up with your own OS of choice you are.
At my place of work, people use whichever OS/Browser works best for the current task at hand... Guess what, they have issues with Chrome/FF/IE/Edge, but just shrug their shoulders and use another that does work because, guess what again, they have work to do.
Someone earlier stated that for PDFs you should always use an external piece of software, I agree totally. I have mostly made people understand at my workplace that if they need to print a PDF then they should save it and open it in one of the reader programs.
Now... OS's, they are all decent enough for work and personal home usage. I have multiple linux servers, Windows PCs, and MacOS here. I have Android on a Samsung S5 and an iPhone 7+. I use whatever I need at whatever time it is needed. I do not criticize OS's because I know that none of them are perfect. Why can't more people understand that they are all imperfect?
I rarely post on El'Reg, but the amount of negativity from fanbois towareds whichever thing is currently hated to be very disappointing.. C'mon some people, get a life, their are more important things to shout about!!!!!
I rarely post on El'Reg, but the amount of negativity from fanbois towareds whichever thing is currently hated to be very disappointing.. C'mon some people, get a life, their are more important things to shout about!!!!!
If this was a normal bug I'd agree with you, but something that sits in the data stream and is changing information before it gets recorded with permanence is NOT trivial. Someone referred to the Intel coprocessor bug, and I think this is indeed as serious and potentially dangerous, and by blocking any alternatives MS prevents any remediation until they get around to it in their own sweet time.
I'm not normal one to wave the FOSS banner around (as I hate the associated debates) but this IS the exact reason why openness often works better - I have seen Linux bugs fixed only hours after discovery because someone could grab the code, work out a way to address the issue and then submit their solution.
No, this is not fanboi level commenting on something trivial, this really is BAD as in "bloody well stop using this for anything with numbers or engineering until it's fixed and verified" bad.
"Why would you class this as important as the Pentium bug?
It's a bug that prints numbers.... Shouldn't you check numbers when printed?"
So someone should individually check every digit of every number in that spreadsheet / online bank statement / online receipt every time they save to a file?
I'd say this is a lot worse.
Just last month I had saved my annual earnings tax statement (W2 in the US), which is only available online, and what I saw online was completely different from what I saved for backup and printed (different issue, stupid special font problem, lots of "?"'s instead of real numbers printing from any other machine).
I couldn't file online without making up a number for an empty field, and if I tried to file that paper it would have been rejected. If it hadn't been such a large screw-up, but just a couple of digits it could have been a huge hassle. So many things wrong in that sequence, but anyway.
So, you didnt check? So you relied on another company?
I couldn't file online without making up a number for an empty field, and if I tried to file that paper it would have been rejected. If it hadn't been such a large screw-up, but just a couple of digits it could have been a huge hassle. So many things wrong in that sequence, but anyway.
..is half-broken. From the idiotic message "If you set this as your default printer, Windows will stop managing it." which does nothing but confuse most people, to the way printers added from a print server show up weirdly for a while before settling down. (after the normal insane delay just displaying devices and printers) Oh, and drivers are missing for very common printers too.
Then there's the actual printing--why is it printing generally works the worst from Microsoft's own apps? I was trying to print out a BitLocker recovery key using Notepad FFS and got nothing but blank pages. Word did something else odd. Pasting into Notepad++ allowed a normal print. Print drivers? Well maybe, but this was printing to a very common HP Laserjet using MS's own drivers, and doesn't explain well why it works in some apps and not others.
I'd have to agree that printing seems to be like something Microsoft's devs slapped on after waking up with a hangover the morning of the release date and saying "Oh shit dude, we forgot to do printing!"
"I'd have to agree that printing seems to be like something Microsoft's devs slapped on after waking up with a hangover the morning of the release date and saying "Oh shit dude, we forgot to do printing!""I have relatively few printing problems in Windows, I suspect because I have a Postscript printer. At least in Windows I can choose whether to print in colour or monochrome in the Print dialog box. Cinnamon Mint requires opening a web browser to choose the correct setting at the printer. There is no option in the Linux Print dialog box to do this.
"Cinnamon Mint requires opening a web browser to choose the correct setting at the printer. There is no option in the Linux Print dialog box to do this."
Really? I run Mint and I can choose color or b/w. Admittedly cancelling a bunged up print job is a little harder in Mint, though it at least cancels quicker than in Windows..
"There is no option in the Linux Print dialog box to [print in monochrome or color]."
Unless I'm totally misunderstanding you, that's not the case under MATE, TDE or KDE 4 in PCLinuxOS -- it's offering me 'Color' options of Vivid, Natural, or None. I do have to manually click on my printer (even though it's the default) to see that tab, though; maybe Mint/Cinnamon have the same quirk?
@ Unicornpiss & Trilkhai
Print dialog box in Firefox allows choosing a printer, paper size and orientation. That's it. No duplex option even.
Chrome allows duplexing and access to a further dialog box where I can adjust the amount of cyan, magenta, yellow and black individually. But then I need to remember what the original settings were to shift back to colour printing.
On Windows 7 Chrome's print dialog has a drop-down for Print in Black & white, or Colour. Firefox's print dialog has a checkbox to print colour images and backgrounds.
Cinnamon Mint 17.3, Lexmark C543dn.
I've noticed this for years (but never seen it reported) ... when sending from Outlook (2010 in my case) the send/receive dialogue numbers are wrong.
If Outlook is sending 1 message, the dialogue says "Sending message 10 of 10"
If there are two messages in the Outbox, it says "Sending message 10 of 11" and then "Sending message 11 of 11"
So, Microsoft programmers have previous for not being able to do sums.
I'm so glad I've found something no one else seems to have heard of!
Out of curiosity, I've just looked at my partner's Outlook (also 2010). When she sends one email, the dialogue says "Sending message 7 of 7"
Weirdly inconsistent.
Both PCs running Win7 Pro