Hacker News new | past | comments | ask | show | jobs | submit login
What happens when placeholder text doesn’t get replaced (elezea.com)
246 points by pascal07 on Feb 24, 2014 | hide | past | favorite | 94 comments



A recent up roar occurred in the world of chemistry when a paper was published with the phrase:

"Emma, please insert NMR data here! where are they? and for this compound, just make up an elemental analysis…"

Here is a link the supplementary information where the quote is located on page 12. http://pubs.acs.org/doi/suppl/10.1021/om4000067/suppl_file/o...


We had a very funny example of this at my school paper. Our football team was one game away from a winless season, and many people were demanding that the head coach be fired for this terrible performance[0].

As a joke, someone referred to Coach Wilson on a pull quote as "head coach of the Columbia football team until Monday morning". This was supposed to be removed before print, but nobody caught it.

To make matters worse, the coach actually was fired that weekend (not even 24 hours after the team had its only win of the season)! That led to an incredibly awkward retraction: ("This was a joke, and we're sorry... but apparently we were right, even though we didn't know it"[1])

This sort of stuff goes on in newsrooms all the time. I'm actually surprised that these mistakes don't happen more often, given how common these are, and given that most copy-editing happens in the wee hours of the morning, fueled by caffeine and sleep deprivation.

[0] Don't have time to scan through the PDFs of the printed versions (IIRC it was only in the printed version), but it was one of these articles: http://www.columbiaspectator.com/tags/norries-wilson

[1] I was on the board of the paper at the time, so I can confirm that we actually didn't have prior knowledge of this - it really was just a very amusingly-timed joke.


It goes on in student newsrooms all the time, but the only time I've ever seen it in real newsrooms is from recently hired students.

It's much too easy a mistake to make, and in some places using "real" text is a discipline offence, so it's much more common to see "cgclcgl" or "hdyhdyhdy" or "123123" or eye-catching text like that. (Lorem Ipsum doesn't leap out at the eye enough).

I once wrote a headline about an invasion of a new breed of octopuses coming to our waters, along the lines of:

"Indian Octopus

Heading Here"

but seeing it on proofs freaked everyone out too much and it had to be changed.


Check out the correction here: http://www.nytimes.com/2014/02/23/nyregion/the-once-and-futu...

Because of an editing error, an earlier version of this article misstated the name of a website that reported on the God’s Only Demons Motorcycle Club. It is DNAinfo, not DNAdisinfo.

And that's the New York Times.


It definitely does happen in newsrooms of journals of record as well. I'm not going to call any out by name, because I know it's very common. They're just generally better about catching it.


Why doesn't print software include a "placeholder" text formatting function? You highlight some text and mark it as placeholder. You can toggle on highlighting when you want to find it to replace, or toggle off to see what it looks like as a final piece. It could also auto-generate lipsum for you. Also, before you publish it, it'll warn you if you have any placeholder text left.


Adobe InDesign certainly does support this. Commercial newspapers are usually using some archaic program though.


Having seen the software used by many newspapers I'm not surprised they don't. But this sounds like a great idea to me.


it's not that they don't necessarily have the functionality- maybe they just don't use it?


Why would there be a retraction for that?


One of the very best one I've seen back when I was working in the book publishing business (many moons ago) was this: a bogus screenshot (totally unrelated to the book) with a legend saying something like this: "Insert something similar to what's in book X at page 237"... With "book X" being the name of a book by a competitor.

Somehow everybody (proofreader(s), author(s), etc.) ended missing that and the bogus screenshot and that legend made it to the final, printed, version of the book : )


Ouch, that one is painfully embarrassing.


Someone I used to work with once left a alert('W.A.N.K') in the codebase. This was in a fairly rarely occurring branch of code and made it all the way to production.

I've taken that as a warning and never write anything I wouldn't be happy for a customer to see in logs, test-data, comments (although handfuls of sarcasm are still acceptable) or debugging code.


I've taken that as a warning and never write anything I wouldn't be happy for a customer to see in logs, test-data, comments (although handfuls of sarcasm are still acceptable) or debugging code.

Very good advice!

I once worked with someone who had the unfortunate habit of naming debugging log files "shit".

After a server migration, a couple of paths had been incorrectly set, leading to the client complaining about a now-memorable error: "Cannot open: shit".

Whoops


> "Cannot open: shit".

I'm adopting this error message for I/O from now on.


I worked on some software that automatically generated documentation for engineers. During testing someone put "hello fuckers" in the code so that he could easily see when it came out the other end.

Naturally that code was left in and went into production, and the customer rings up to complain they'd just printed out a 2000 page manual with "hello fuckers" at the end of every line.

I've never seen a manager look so furious, but we were all paid so little we just laughed.


> but we were all paid so little we just laughed.

that summarizes the whole thread.


This. I didn't even need to learn this the hard way, just had to read back some of my earliest teenage code (obviously not for any type of "production" setting) and hear enough of these scary stories to decide once and for all, it's really not that hard or big of a deal to be professional about these kinds of things.

Same goes for code comments. Unless it's a really, really good joke that won't offend if it happens to show up anywhere, or over your shoulder.


I was officially forbidden in my old job to write error messages at 3AM. Code was ok. But text - not after а high ranking client saw an error message. Its good that during the meetings with him we both were (in)appropriately vulgar and offensive so only my manager was shocked.


How I regret not watermarking the Lorem Ipsum wine bottle image.

https://mobile.twitter.com/jgrahamc/status/14082734075741798...


You could be raking in the backlinks. http://creativecommons.org/licenses/by/2.0/


I may be an idiot, but I feel like I need to get a bottle of that to keep on my desk. But googling it only seems to reveal a different style of label. [1][2] Have I got the search right? You didn't happen to buy it online?

[1] https://www.google.com/search?q=2009+Roland+Tissier+Sancerre...

[2] http://bestbuyliquors.com/roland-tissier-fils-sancerre-2009....

Edit: looks like different labels for US and UK? Oh well. :( [3]

[3] http://www.colombierwines.co.uk/products/Sancerre-Rouge%2C-D...

OK, enough wasting of time...



Stopping for 30 seconds to come up with subpar copy is usually worth it vs using some sort of placeholder that you need to come back and fix later. When I know the copy is awful and that I need to come back and replace it, I put in a comment that says

  # COPY_REQUIRED
As long as I'm consistent about the tag that I use, I can easily search my project for my tags and fix before launch.

Similarly, I do a fair amount of presentations, and I always put in huge red letters a note to update the metrics, provide a source, etc. It's hard to miss when doing a final flip-through.


When writing papers in grad school, we'd leave placeholders in all the time, but always prefixed with the string "ZZZ", so they'd be really easy to search for.


I sometimes use similar approach with code to indicate really hacky code that should be removed before any bigger releases.


I think that's what //FIXME: is for. TODO and FIXME are both commonly supported, as well as a few others.


Similarly, when adding debugging-aid print statements, I prefix them with ">>>", and make sure I don't use it elsewhere.


why not "debug: "?


There was a great example of this with QR codes, though amusingly I can't find the link at all at the moment...

For a while the first hit in Google's image search for QR Code was a code for a person's personal site which he'd put up as a test and by happenstance got picked up by a few places which a few other places picked it up from and so forth. All fine when you are explaining QR codes and are fine pushing people to some random but currently inoffensive site, but a number of places used it up as a placeholder in adversing campaigns then forgot to replace it with one representing the real URL before going to print.

The owner of the page knew what had happened as he had offers to rent or buy the URL, demands that he turn it over when he didn't want to, and even threats of being sued for "hacking our advertising". IIRC he never "sold the URL" instead leaving his personal page in place. He had the good grace not to add something offensive to the page, which I'm sure I would have done after receiving any such demand or threat. I would have least published a list of the fools that fell into the "trap" with a copy of relevant correspondence (sanitised to remove names and such).


This sort of thing can also happen when people are not really able to check to make sure the content is correct. For example, see http://www.theguardian.com/theguardian/2008/nov/01/5.

Fortunately the text was bilingual and I doubt there are many Welsh who don't speak English...



What's the story behind this? Some common shopping cart software?


Yeah, GSI Commerce (now Ebay Enterprise) I think.


Such an instance lead to a libel lawsuit at Virginia Tech in 1996 when the student paper left "director of butt licking" as the title for an administrator in a pulled quote.

Article from '96 on the filing: http://www.splc.org/news/report_detail.asp?id=41&edition=16

Article from '97 when the judge dismissed the case as it was "not defamatory": http://www.splc.org/news/report_detail.asp?id=136&edition=14


That's why I always use "_X_" as placeholder text. At the very end of development I run a search for "_X_" on the templates/files to see if I accidentally left any. Not only does it stand out visually but it's very easy to run a search on without any false positives.


An ATM I use frequently hasn't been properly set up, so during the transaction, you'll see things like "[CONSUMER MESSAGE]" and "[NON-MEMBER FEE]" and "[EXIT MESSAGE HERE]". Makes me smile every time.


It's funny but I think that's the way to go if you can't get your content first, if something slips through, it might be weird to the customer, but not insulting.


In this case it's designed this way because the ATM vendor is not the same company as the bank that owns/runs the ATM. The bank will want to put different text depending on the locale, default language, their own bank marketing jargon, etc. This is a perfect example where design should come before copy.


I know someone who used to work as a programmer in finance. One of his coworkers liked to use creative placeholders in his email templates. One day something went wrong with the placeholder substitution, and a bunch of their clients received emails starting, "Dear ${rich_fat_bastard}". That guy's career in finance didn't last too long.


I learned this at 18, when after a particularly frustrating round of edits on some file, I sent it to my partner for review with a filename like "fuckthiscustomer". He then forwarded it on, without reviewing.

Customer wasn't pleased and demanded an explanation. Somehow my partner convinced them it was an email virus in their system. Still, lesson learned.


So good -- I remember seeing one in the NYT when the George Zimmerman jury verdict was reached, that was just a XX/XX/XXXX in the date section. Assuming they had both articles written.

I suspect this will only become more and more common in the "publish first, edit later" evolving world of online journalism.


It's been standard practice since long before the Web to have pre-written obituaries for high profile people (politicians, royals, the Pope, etc.) and to update them periodically. When someone of this stature dies, the newspapers will have multiple pages of content ready to roll-out with minimal editing.

Some people once found a bunch of work-in-progress obits on CNN's website:

http://en.wikipedia.org/wiki/List_of_premature_obituaries#Th...

And the BBC had a 30 min comedy about an obit writer who saved their job by murdering high-profile individuals:

http://www.imdb.com/title/tt0653345/


Reminds me of the SNL clip where Tom Brokaw was going on vacation and had to pre-record every possible news story:

https://www.youtube.com/watch?v=SkhwiuRbOEE


It is known that news organizations already have obits for prominent figures written and ready to be published at a moment's notice. CNN accidentally exposed some of these in 2003.

http://www.cbsnews.com/news/cnn-chagrined-over-premature-obi...


The other way to look at it is that it is the logical way to go about optimizing delivery times. Parallelize everything you can. I don't see why people are shitting on it, it's not like you'll wait for the designer to pretty up your web app before you can implement the functionality you want.


I worked at a place (but not on the team) where someone used the Ebonics (http://en.wikipedia.org/wiki/African_American_Vernacular_Eng...) version of Lorem Ipsum to fill in a large amount of text on the app. They forgot about it and showed it to the client and it contained a lot f-bombs and n-words.

It seems like people need to learn the hard way to not put that stuff in. If it's there, it'll be accidentally shown to the wrong person.


> and it contained a lot f-bombs and n-words.

Yeah, calling that "Ebonics" is being too charitable.


...And this is why you never put insults in placeholder text!


Recently I've started censoring myself with log message even if I know they can't survive for more than one page refresh. It's tempting to take out some frustration with a "console.log("fuuuuuuuuuu!")" but in the end it's not worth those verrry rare events when you're accidentally insulting a customer.


It was certainly covered in day 1 of my journalists' training.


It's a testament to how much more software we could stand to develop, software that probably hasn't even been touched, yet.

For a newspaper, obviously they need to have a parallel process to layout the paper and develop the content. Why don't they have software now that "compiles" the paper from design files and content files and won't release the copy until all of the content is marked as reviewed?

Or essentially, as the person is performing the layout for stuff like callouts, instead of generating their own placeholder content, they generate rules for the place holder content (just as they seem to be typing rules into the callout as the placeholder itself), and the system would both generate the placeholder for the designer, while also queuing the snippet of content for the writer.

Then, you just have to train your users "never type in your own placeholder text." Use the queue as a project management tool. Editors could then review the text, mark it as reviewed, or re-enqueue it for rewriting. The article is done when the queue is done.

I mean, really, I'm not even describing anything revolutionary here. It's BugZilla, Redmine, etc., just with a layout program tied to the fields in the database.

Would newspapers and magazines actually use something like that? Or is pigheaded entrenchment into old ways the disease of their industry that is leading them to die out?


Sounds like something along the lines of http://en.wikipedia.org/wiki/Adobe_InCopy:

> Once integrated, writers, editors and designers can simultaneously work on the same page; the designer creates the page layout with InDesign, while editors simultaneously edit different stories with InCopy, via the Adobe LiveEdit rights management system.

I recall there's at least one other product for similar kind of workflow (not tied to Adobe ecosystem), but being no longer in the industry I forgot the name.


So is the software any good? Why wouldn't something of such obvious utility take over? I have seen plenty of cases where users will circumvent a system they are given if it is difficult to use or they don't understand it.


I know of one example where this kind of integration is actually in place and properly utilized. It's a pretty big newspaper. The integration is provided by a publishing system called K4.

Admittedly, very few publications would employ similar systems here. Because they're complex to set up, expensive, and not easy to pirate (I don't think many small publications paid for their copies of InDesign/PageMaker), because there's inertia and fear of losing their jobs among the staff (not many young forward-thinking professionals in print nowadays), etc. Not sure how things are US, though.


Okay, not surprising.


It would have helped stop the headline "Can Dec anally match Ant?" in the Daily Express. This was credited as "Accidental headline of the year" by the Guardian.

http://www.theguardian.com/media/mediamonkeyblog/2009/sep/01...


Also funny is when error text makes it through. Back in the '80s, I'd occasionally see in the classified ads section of the Los Angeles Times ads that looked like this: "FILE ID=18273 NOT FOUND".

The Lorem Ipsum text greatly confused me the first time I encountered it. That was in Apple Pages. I made a new document using a template, and it was full of that text. I didn't realize it was meant to be nonsense. I assumed it was meant to be sensible sample text, and should have been English for people in the US, and therefore I must have somehow managed to screw up my internationalization settings somewhere and it was showing me some foreign language's sensible placeholder instead.

I spent a long time trying to figure out how to change my settings to get the Enlish placeholder, before finally doing what I should have done in the first place, and Googling the actual text.


One of the software products I worked on once forgot to take out test data from their reports. We got a support ticket asking what the report "I can hear angels" did.


I'm not convinced that last example wasn't a winemaker looking for some fancy-looking latin to make his wine label look classier.


Nope. I took that photo and tweeted it. I subsequently phoned the vineyard. They had no idea about the mistake.

https://mobile.twitter.com/jgrahamc/status/14082734075741798...


Remarkable effort on your part :D


I was thinking the same. It would even be better if the winemaker was aware of the origins of the text and added it as a typographer in-joke.


Have a look at the rest of the label. Now, you seriously think that?


Humans make mistakes and I'm glad we have those placeholders creating funny situations.

However I don't put anything in my comments / code that would reflect poorly on me or my company. A bit of humor is always refreshing, but vulgarity is a no no.


Insults is definitely a bad idea but think that dummy content is unwise in general. In my experience, actual content almost always makes the design look different from dummy content.


I've also experienced this. Recently I interviewed at a big tech company and when I received the original e-mail it said "Hello NAME" instead of my first name.

I asked if it was recruiter error or programmer error and never got a response.


Probably because your email was internally forwarded to PUT_AN_EMAIL_HERE.


I found Lorem Ipsum on the side of a Chipotle bag not too long ago. I thought it had to have been satire, but now I realize that it's just incompetence.



That is pretty clearly intentional.

It's hand-drawn. It's not something that someone was going to type over later.

Chipotle bags have all kinds of different wacky, rambly messages written on them. It would not at all be out of place for them to say "hey, let's just throw in a Lorem Ipsum, then a few design nerds will be amused by the bag".


On the one hand you should cut people some slack. Content First = noob (okay, that's probably unfair but...)

I've worked on projects with 300 + lines (yea, not many) but that had to be translated into 17 languages. That's 5100 things that need to be checked. I know of games that with all the localization had over 40,000 dialog audio files (yes, that's four zero k).

Content First doesn't work on big projects IMO. In fact with "agile" you won't even know what content you need until you get there.

That said, with a little forethought it would certainly be possible to try to design some system so you could at least kind of auto verify that at least all your placeholder stuff has been replaced with something not placeholder. For example if you prefixed all placeholder text with "PLACEHOLDER: " and then had a script that would check. Could probably do similar things for audio files though it might be a little harder.

Unfortunately for people adding new text or audio if they forgot to follow the rules you'd have a hole. I suppose you could require first check in to follow the placeholder rules? Or maybe someone has a better idea

Now for a project I worked on. If you play the US Crash Team Racing (CTR) on PS1 and unlock Penta Penguin (IIRC), and then race with him you'll hear "test sound 1" and "test sound 2". I don't remember how many sounds there are in CTR but Penta Penguin was only really in the game for the Japanese version which could only be unlocked by going to a promotional event and getting a special save stored on your memory card. So, being a relatively secret character no one play tested him in the USA and no one bothered to go over the 1000s of sound files to see which ones were still placeholders until after it shipped.


I mean, yeah, I'm sure these were all embarrassing at the time, but keeping in mind that newspapers are published every single day, you're talking about an error rate that's extremely low, relatively speaking.

I'm actually not seeing how any of these examples would have been helped by content-first design. Having the content first is just a different approach with its own unique set of issues (namely that you'll be changing the content 700 times during the design process, which is actually a bit nightmarish when you're passing data around). I wouldn't call it better, though.


I was an editor of a small print publication and this happened to me once. The caption below a photo was something like "Please tell me this isn't a stupid picture of XYZ" (can't remember what it was) as a joke to the copy editor who would eventually see it an laugh at how hilarious I was. Of course it slipped by because it just looked like a normal caption.

I'm not sure these examples advance this article's point too far-- incidents like these could have been avoided by more clearly marking that the text was a placeholder.


I saw a padlock for sale with the packaging in English, French, and Spanish. Unfortunately they left Lorem Ipsum in place of some of the translations. This would be harder to spot than most non-replaced placeholders, even if someone is proofreading. (Latin is in the same Romance language family as English and French.)

Photo at http://www.righto.com/2009/11/lorem-ipsue-when-international...


A very recent example: The newsletter announcing the blackphone (discussed here as well).

Upper left corner says:

Use this area to offer a short teaser of your email's content. Text here will be shown in the preview area of some email clients".

https://dl.dropboxusercontent.com/u/1194203/Screenshot_2014-...

Oops :)


Funny enough I just found another one today.

http://www.econstor.eu/bitstream/10419/75793/1/cesifo_wp503.... [pdf]

all the figure placeholder text weren't replaced and the figures are missing.

Figure 1 about here

Figure 2 about here

Figure 3 about here

...



'HED', 'LEDE', 'GRAF', 'DEK' and a number of other specialized spellings are often used for placeholders because they (theoretically) stand out for copy editors and spell checkers.


Ok, I give up. Where is the placeholder text on the first newspaper? The 4th picture down, has "A Push To Let Ryan Be Ryan" as a heading.


Last sentence of that paragraph says "And another line here."

Took me a while to find too.


The last line, where it says "And another line here."


Methinks it may be wise to add grepping for placeholder text as part of one's build process.


Premise is silly. You want deadline-based newspaper designers to WAIT until all the stories and headlines are done - then design the paper. And get a paper out? I take it you internet junkies know nothing of the once enormous scope of journalism and the resource allocation necessary to accomplish it. Let me explain parallel process project management to you someday.


Also, it’s not important. Stuff like that happens, but it happens rarely. Most of time it works and if it doesn’t it usually isn’t a problem. So sometimes there’s silly placeholder text. Who cares?

This is not a fundamental problem that needs to be solved. Sure, it’s easy to develop some better guidelines on how to handle this (e.g. guidelines that make sure potentially insulting language is never used and make all placeholders easier to find in an automated way) but that’s about it. There is no need to fundamentally change how production works to solve this.


That enormous scope is obviously what is killing the newspapers.


What is killing the newspapers is the much cheaper and much better targeted ads delivery via the internet and smartphones.


Dunno if smartphones are really in the picture (yet), but between competing with TV for ad dollars and Craigslist for classified ads, they are in a tough spot.


What is killing newspapers is that paper is no longer the best & fastest way of getting news now that we have newscreens.


Papers are quite capable of doing this sort of thing without deadline pressure. The Mirror, for example, once announced to the world that "Monty Flies Back to Front" (1942), and in 1952 told their readers that "Sir Vivienne Fuchs Off To North Pole".


My personal favorite: "Foot Heads Arms Body"


Fortunately we have blogs instead! /s


This is a reminder to all the junior developers (and some of the forgetful senior developers) as to why you never put anything in testing text that you don't want the world to see. Having users see "Fucking FuckBalls" as placeholder instead of "John Doe" is alarming and will lead to quite the awkward discussion with your boss.


first job in early 90s doing website for huge newspaper. fill elections page with outrageous but real quotes from the campaign. editor assumes a few are good to go and publishes. major shitstorm. team makes a standard to only use ipsum lorem or Nononono.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: