I own several icon sets but the one that I return to over and over is Vincent le Moign’s copius Streamline pack, which almost always seems to have just the pictogram I need when I dig into its catalog. In fact, when I published my Design Tools Survey recently it was Streamline I turned to when I needed to illustrate it quickly.
When I wrote about Streamline a year ago, the downloadable pack shipped with an impressive 1,600 icons. Now le Moign has released version 2.0, which weighs in at a staggering 4,000 icons. To be clear, that’s four thousand original drawings; when you figure in the multiple sizes and file formats of each icon that ships with the pack, the total number balloons even more.
The sheer volume of labor required for a project of this size seems as if it would require at least five or six designers if not many more, but when I asked le Moign over email about Streamline’s new version, he insisted that he was able to complete this revamp of the set (which includes improvements to many previously issued icons) with just one additional designer and an assistant. The dedication required to do this, not to mention the repetitive strain injury, boggles the mind.
With that many icons it’s probably true that Streamline is one of a few outliers, but it’s not uncommon to see icon packs with hundreds if not thousands of icons these days. Is this an arms race? Are icon designers escalating their numbers purely because that’s what drives sales? When asked about this, le Moign admits, “Big numbers impress and help the sales, sure. But it’s also in the interest of the designers who use these packs. With smaller packs, you may think you don’t need that many icons until one day you need one that’s not included. I regularly get emails from customers asking for more icons covering more subjects, like hotels, real estate, marketing, etc. It makes a pack much more valuable to know that you have a great chance of finding a new icon when you need it, and that it’ll fit stylistically.”
Having access to four thousand icons is of limited use though if finding them is a chore. Downloadable icon packs, in particular, are notorious for being organized according to unique schema. One publisher may issue everything in a single file, while another may distribute each icon as its own file; some sets may be parsed out into folders while others may require a legend in PDF or HTML format. Even the most thoughtfully prepared of these sets is laborious to search at best.
All of which sounds like a great argument for the terrific work that The Noun Project has done in creating a massive online repository of commercially available icons. But the continued proliferation of sets like Streamline seems to argue that lots of icon customers are comfortable with downloading these packs. For his part, le Moign does foresee standardization just over the horizon. “Searchability is the biggest request from customers,” he says, “and that will come from a standardized way of distributing these assets.” He’s already made Streamline compatible with IconJar, a burgeoning, iTunes-like desktop application for organizing and managing icons based on a new file format—.icnjar—that its creators hope will become a standard for icon distribution. Le Moign himself is actually also working on Iconbook, an open source project with similar aims.
Once searchability is solved, is there a practical limit to these icon sets? If Streamline can grow 250% in just a year or two, how far off is 10,000 icons—or more? “Ten thousand sounds crazy now,” he says. “But when I started more than just one thousand sounded like a fantasy.” And yet here we are.