« Campaign Flowchart | Chocolate Direct Mail »
Frank Romano created a tsunami in the industry earlier this year on a PrintCEO Blog when he attacked suppliers for not standardising on PPML/VDX as a variable information (VI) language. Opinions on this podcast have been widely shared since on industry discussion forums, however, I thought that it would be worth revisiting this topic and draw out the facts from fiction.
It's no secret that Frank is a long-time advocate of PPML/VDX and was part of the committee responsible for establishing this language as a standard; originally an ANSI standard in 2002, PPML/VDX is now an ISO standard (ISO 16612). So, naturally, Frank would speak favourably about PPML/VDX, but he goes beyond that and makes some pretty alarming claims.
In his podcast, Frank explains that the reason that "only 10% of all digital printing involves VDP" is because the "suppliers have screwed the system up" by not adopting PPML/VDX as a standard. Now, before I launch into a 'Frank attack', I would like to point out that I have great respect for Frank's opinions and recognise him as an industry visionary, but in this instance, I strongly disagree with his comments.
Frank could be correct that "only 10% of all digital printing involves VDP", I don't have any information to claim otherwise. However, this statistic is hardly surprising as VDP is still in an evolutionary cycle — it has grown considerably over the past five years, but it's growth has been fairly proportional to the rise in other digital print volume, such as offset transfer, which has been enabled by faster printers, higher quality print and affordability. To make a claim that suppliers have screwed the system up by not adopting PPML/VDX as a standard is a peculiar one.
As Frank correctly points out, only two digital print vendors support the PPML/VDX "standard" (Kodak and Xeikon). The reality is that this probably will not change, it's unlikely that other vendors will adopt PPML/VDX — it may be an ISO standard, but it arrived a little too late. When digital color printing emerged in the early 1990's, the legacy data stream languages used for mono transactional printing were not suitable for personalised colour documents, so as a result printer and RIP vendors developed their own VI languages to enable print engines to run at rated speed. In 2000 PODi developed PPML. So, from early on, vendors already had support for their own VI languages and many chose to later add support for PPML. PPML works — the industry really doesn't need another VI language. Or does it? Let's examine the technical merits of PPML vs. PPML/VDX...
PPML is an XML-based language. The PPML framework is built on two core methods, object-level granularity (which describes content objects on one page) and reusable content (the ability to store digital assets to the RIP memory for reuse). Like several other VI languages, PPML can dynamically merge objects (text and images) on a template at the RIP from the supplied dataset and digital assets.
PPML/VDX (formerly VDX) is based on a subset of the PPML specification. It's a PDF-based standard that uses a subset of PPML to define the reusable content within the PDF file. A PPML/VDX file can consist of one or more files, but in its most basic form, it will always contain a layout file that is essentially a PDF (but uses the extension .vdx). The layout file functions as a container for the VDP template and variable elements (text and images) and also includes PPML information that defines the document layout.
So, PPML/VDX is actually based on PPML, so what makes it better than PPML? Well lets examine Frank's claims about PPML in turn:
1. "It's not a standard"
True. PPML is not an international standard, but it's based on XML (which is a standard). But does it matter whether it's a standard or not? PPML is definitely not proprietary; the PPML specification is widely available to PODi members.
2. "It really doesn't work"
False. I know of many printers who use PPML and I've personally used PPML across a number of different RIPs and created PPML files from a number of different VDP software products without issues.
3. "It has no way of proofing"
False. Xeikon developed a free PPML viewer for previewing PPML files.
In the podcast, Frank goes on to explain that unless suppliers "standardise [on a VI common language], we are not going to see the growth in VDP". I really don't agree with this statement either. At the end of the day (or shift), it really doesn't matter what VI language print providers use; proprietary, open or standardised — as long as they can run their printers at rated-speed. The majority of VDP software products emit a number of different VI languages, so standardisation on a single VI language is not really essential.
I think the real issue here is not PPML/VDX vs. PPML, but Frank Romano vs. PODi. The reality is that PODi won the battle years ago and PPML is here to stay. Sorry Frank.
Posted on Saturday, 6 October 2007 at 7:49 PM | TrackBack: http://www.veedeepee.com/cgi-bin/mt-tb.cgi/34
Listed below are links to weblogs that reference Frank Attack:
» New VDP Standard from veedeepee
ISO Technical Committee 130 are currently working on a new standard for 'variable printing data exchange' in PDF; ISO/WD 16612-2. The actual standard name is not finalized; The working name is VPDF/X-200X, although this will most likely change before t... Read More »
Tracked on Saturday, 17 November 2007 at 8:42 AM
I never argue with my former students -- hopefully they have been educated to challenge everything we stodgy professors say. All of this will be academic within two years. Mark my words and Eliot's.
Frank
Posted by Frank Romano on Tuesday, 9 October 2007 at 4:18 AM
"PODi won the battle long ago" is interesting commentary when many of the case studies that PODi has showcased have not been PPML. When you look at the last 7 years worth of PODi case studies you've got everything from Enhanced Postscript, to VIPP, to VPS, to other weird and wonderful proprietary formats.
My experience in the field has been less optimistic than yours Eliot... my experience has been that not all PPML is created equal... for the longest time from a multitude of different emitters PPML has worked on one RIP and not another... worked out of one package and not another. I think mostly this is because no one has really gotten behind PPML which is mostly what I think Frank is saying - however the "Why?" question as in "Why haven't vendors gotten behind it is probably much more complex?"
I've only been witness to the last 20 odd years worth of variable data printing - but what I'm left with is definitely the observation that Variable Data is definitely not new... I can't go back as far as Frank, but I can remember the early 80's when SGML which gave rise to XICS which even at that time was capable of variable composition in full colour.
The problem with Variable Data is that it is not mainstream enough to have evolved mass standardisation - what I mean by this is that it's always been a bit of black art.
I can remember doing VI on Xerox' first commercial colour laser printer the Xerox 4700. It had a postscript 1.0 compatible RIP written by Xerox and we had VIPP loaded on it. This was late 80's early 90's era and VI in full colour with more than just simple graphics and text variable caching was more than just possible we had live applications at customer sites.
Again, the problem was that Variable Data was hard (a black art)... and therefore it was up to the vendors, like Xerox, HP, IBM and Kodak to invent the technology that made variable data possible within the realm mainstream technology.
Now... in simply remembering this I also remember when there was still a divide in the industry on Page Layout tools, Page Composition vs. Page Description Languages vs. Printer Command Languages vs. Machine Code and Dot Addressing. Add to this the fact that there was fierce competition among all the major print vendors, no wonder things evolved as they did.
One thing I will say... it is a bit simple to say that the VI standards should have been adopted decades ago when the industry couldn't even agree on standards for PDL, page composition, etc. There was also the issue of hardware and RIP platforms getting faster every single year - for some vendors the problem of variable data would eventually go away because you'd have RIPs so fast that even with fully composed print streams you'd always be ahead of the printers.
Whoa, now 20 years later... and with engines that print almost 50-1000 or more impressions per minute in full colour and we're back to wondering how we can optimise the datastream to keep the engines busy.
The only thing I notice 20 years later is that Variable Data is less of a black art (easier now than it has ever been)... there's even more choice by way of VI tools and Page Composition, Postscript, PDF and Adobe dominate the scene... and still most people have no idea of what VI, VDP, or Variable Data even are... and I guess the question I have is should they??
Let's hope both Frank and Eliot are both right and all this will become academic in a few years and nothing more than history and a memory for some.
Posted by K. Duane Mackey on Friday, 12 October 2007 at 9:49 PM
I agree Duane. I'd like to clarify that PODi have not won a VI language battle, but PPML has succeeded over it's near (and somewhat intentional) rival; PPML/VDX. As to the future, I'm sure looking forward to it.
Posted by Eliot Harper on Friday, 12 October 2007 at 10:13 PM
I somewhat agree with Frank, although having worked with many different vendor solutions over the last ten years, I don't think is just the number of options or standards.
Most print vendors simply do a poor job of educating and training users when it comes to VI software. Not the sales concepts, but the task of expertly using the software.
Print system suppliers especially still live in the 'box drop' world where hardware, not software is king. Software solutions are pushed at trade shows and in fancy powerpoint presentations. The intent is there, but little resource or budget ever allowed for good user training. From discussions with colleagues worldwide, this isn't true everywhere, just in 80% of cases.
There's also the fact that not all graphic design people brought up on Macs are comfortable with VI, since variable print is very much a process, not creative chore. Sometimes new people with an IT backgrounds are needed and few vendors will warn the client of this possibility since it could affect the sale outcome.
Cynically yours....
Posted by Kevin Trye on Friday, 23 May 2008 at 1:36 PM