Disqus - Latest Comments for harrisj

Re: Solving a Century-Old Typographical Mystery

harrisj — Fri, 27 May 2016 08:20:50 -0000

Thank you for the explanation. I will admit I have only a passing knowledge of various types of printing so this is good to learn.

Re: Solving a Century-Old Typographical Mystery

harrisj — Fri, 27 May 2016 08:19:57 -0000

Indeed, I struggled about using a modern term as a hook to draw people in, even though as you and others have noted, this wasn't strictly ASCII art or even Baudot art since it didn't involve any character encoding and likely wasn't even typewriter art for many of these examples. I used to use a typewriter myself, so I totally understand how maddening how it could be, although the ability to turn the roll in sub-line increments and overtype existing letters allowed some interesting work not possible with the strict spacing of computer terminals

Re: Solving a Century-Old Typographical Mystery

harrisj — Thu, 26 May 2016 15:34:30 -0000

You are correct, although the salient difference for me between concrete poetry and these decorations is that the text of a poem makes sense even if the formatting is removed, but still a similar style

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 22:35:57 -0000

You are right that I can't say it was Hot Type exactly, and even in the case of the Brooklyn Daily Eagle, they don't seem to have moved to hot-metal typesetting in 1892, but to a larger composing room with better tools. Still, that might have been enough.

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 22:32:04 -0000

This is very interesting, because I had naively assumed that since the final product was on the printed page and it predated the arrival of hot lead typesetting, that it was all done by hand (in negative form essentially) and that a typewriter would not have been involved. But it seems like that assumption was mistaken if I understand you correctly.

I do also indeed appreciate the artistry and style of the Brooklyn Furniture Company ads – perhaps because they undermine some common conventions of newspaper advertising – but what I find most mystifying is that no other advertiser seems to have followed suit. Perhaps I am just looking in the wrong places, but they really seem to have occupied a unique place in the advertising universe in NYC.

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 22:27:48 -0000

You are very right in that it could be a mask applied, but I too feel like offset lithography wasn't in wide use then. I suppose it could just be a mask over the block of text on the page itself, but that would have to be durable. But I will admit I know little about the processes back then.

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 15:13:23 -0000

This is amazing and I know what I will be flipping through this evening!

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 15:12:51 -0000

That is true. It seems that the very first forays were just arranging text into shapes like the hoop skirt example I had. But then the next steps were figuring out that they could make larger letters and other abstract shapes.

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 15:11:17 -0000

Yes, it seems like the two are related, although as always I could be missing some cases where the two happily existed side-by-side. I had assumed it was because it freed up compositor resources for advertisers and made custom ads cheaper, but that is an interesting point that the interface was different for compositors too.

Re: Solving a Century-Old Typographical Mystery

harrisj — Tue, 24 May 2016 07:56:52 -0000

Exactly, although it did involve someone forming the original block of type which was either etched or masked. But it's so mysterious in that it is different from the other ads and I haven't found that it ran anywhere else yet. Why? And why did nobody else besides the Brooklyn Furniture Company do more with type art than making big letters and basic iconography? I likely will never know.

Re: Solving a Century-Old Typographical Mystery

harrisj — Mon, 23 May 2016 22:26:34 -0000

I wish I could be there and find out who created these ads, because I am dying to know. The Brooklyn Furniture Company seems to have had a pretty inventive advertising department, but I also see them collaborating with printers at newspapers who might've encountered this technique elsewhere.

Re: Solving a Century-Old Typographical Mystery

harrisj — Mon, 23 May 2016 22:24:56 -0000

Well, I hadn't had my coffee...

Re: Solving a Century-Old Typographical Mystery

harrisj — Mon, 23 May 2016 17:13:36 -0000

I apologize for that. I will admit that I was originally shocked but I suppose that the ingenuity of printers should never really surprise me. I call it ASCII art-style as a reference although of course it isn't ASCII or even Baudot character encodings being used here, but because it's the style people know. I still found it surprising that this wasn't art per se nor created on a typewriter, unlike other forms of text art.

Re: Solving a Century-Old Typographical Mystery

harrisj — Mon, 23 May 2016 17:10:06 -0000

Nice! I can't even find it on Google (which keeps trying to correct to Boredom). Do you have a reference handy?

Re: Solving a Century-Old Typographical Mystery

harrisj — Mon, 23 May 2016 17:09:34 -0000

This is a good point. I will admit that I am freely speculating here because I haven't yet found any good descriptions of advertising tactics in the late 19th century (I have no doubt those exist, but the search is difficult). But you are right that newspapers might potentially intervene to prevent a fully-unrestricted evolutionary battle ground and this might've had effects distributing ads across multiple newspapers. Thanks!

Re: Why I Like to Instagram the Sky

harrisj — Tue, 15 Mar 2016 07:40:38 -0000

Thank you! I too am a jaded Gen-Xer and I am a bit confused why my detractors seem to assume I am a Millennial hipster. Referencing an art-film movement from the 90s seems like an obvious giveaway that I am older than that. I guess it's a reflection of their stereotypes about Instagram and selfies.

Re: Why I Like to Instagram the Sky

harrisj — Mon, 14 Mar 2016 17:24:00 -0000

Thank you for the kind words (and also for not assuming I am some sort of bearded hipster like the commenters above). I would love to see what you do with this or any other artistic exercise in constraints.

Re: Chatting with bots: How Slack is changing how newsrooms talk amongst themselves

harrisj — Wed, 11 Jun 2014 13:19:02 -0000

Fair enough. I'm just noting what we use it for. I think secure group chat is an interesting problem, since most of the solutions I've seen usually fail in some way because they rely on a shared key or something that falls apart quickly if compromised. Otherwise, the UX is a nightmare because things like search are not supported. I don't know what the best solution is out there.

Re: Chatting with bots: How Slack is changing how newsrooms talk amongst themselves

harrisj — Wed, 11 Jun 2014 11:58:47 -0000

It's funny because when I look back at our own Slack usage at the NYT, we are using it solely for the devops needs that happen to be a large portion of our email and point-to-point IMs. For instance, the election loading examples you see here. I don't think we would use it to discuss sensitive stuff about sources for these very reasons.

I get where you are coming from here, but I don't think that group chat like that should be rejected because it is not completely secure. In general, i think of these problems like there is a slider on a control panel with Convenience on one side and Security on the other. I accept that there is a risk of my messages being eavesdropped, but I'm willing to accept that for these specific use cases.

Re: Distrust Your Data - Learning - Source: An OpenNews project

harrisj — Fri, 30 May 2014 08:52:46 -0000

Exactly. I think there is a tendency in modern business to equate something being quantified with something being measured, even though we are usually making errors in our measurements and we are measuring a proxy for what we really want. These issues are probably even more widespread in the world of Big Data

Re: Distrust Your Data - Learning - Source: An OpenNews project

harrisj — Fri, 30 May 2014 08:51:29 -0000

Excellent comment. I've been thinking about this subject a lot since I read that ebook on data skepticism I linked above. The author is a data scientist in the financial industry, so she's seen a lot of bad cases where bad data was trusted or shoddy proxies were applied. The illusion of precision is a vexing one indeed. Nothing drives me crazier than when I see an average quoted to 5 or 6 decimal points as if the raw data were that exact...

I would love to know how to better parameterize the contamination of the data. I usually have a rough idea of how good or bad a data set is, but nothing that translates into anything as precise as a margin of error

Re: And Remember, This One's for Posterity - Learning - Source: An OpenNews project

harrisj — Thu, 14 Nov 2013 14:15:46 -0000

I wrote this piece before I cohosted a discussion on this subject at Newsfoo. I'm not the only one who has been troubled by this. Several people noted that the site http://chicagocrime.org that inspired them to make news apps of their own is long gone, because the creator folded it into Everyblock and then NBC shut down Everyblock. Furthermore, since it was dependent on a hack of a specific version of Google Maps, running the code today might not work. Google image search returns screenshots, but those are scattered around the web too.

One intriguing idea was that we could standardize some image / screencast approaches for capturing the functionality of an application, making compliance part of the application process for some journalism awards (we could also prioritize previous award winners for preservation first). Alternatively, it should be possible to create sandboxes that represent specific browsers and certain points in the Internet. If you could preserve a web archive with all your external dependencies, then you could theoretically run 2010 Chrome on your 2067 quantum laptop and it would all work provided the sandboxes were updated to still run on that hardware.

Re: The Times Regrets the Programmer Error - Learning - Source: An OpenNews project

harrisj — Mon, 23 Sep 2013 09:42:48 -0000

You raise a very good point. I spent so much time talking about how I might mess up data, I didn't even go into how the data itself is likely filled with errors of all types. There are some great stories in that sometimes, especially if the data is supposed to be used for enforcement by government agencies (see recent stories on OSHA for instance) or is touted as a metric (see many stories on crime data or school test scores). Sussing out the quality of the data is step 0.

Re: The Perils of Polling Twitter - Learning - Source: An OpenNews project

harrisj — Fri, 19 Jul 2013 16:36:03 -0000

Also, I'm not critiquing the Super Bowl tweets graphic at the beginning of the piece. I think it portrays exactly what it is supposed to be and serves as a reminder how little we have advanced in visualizing Twitter since it was published.

Re: The Perils of Polling Twitter - Learning - Source: An OpenNews project

harrisj — Fri, 19 Jul 2013 16:29:16 -0000

Well, I mainly just want to present it as my personal opinion. It really depends on your comfort threshold.