RSS
Twitter
IBM Patents Removing Leading/Trailing Blanks
from the _um_-_wow_ dept
theodp writes "With its example of how ' John Doe ' could be saved in a database as 'John Doe' (i.e., without leading or trailing blanks), purported patent reformer IBM dazzled the USPTO enough to earn Big Blue a patent last Tuesday for Automatically removing leading and trailing space characters from data being entered into a database system . The three IBM 'inventors' are also seeking a related patent for Retrieving data from a database system without leading and trailing space characters. Hey, if the patent system ain't broke, don't fix it!"
Reader Comments
(Flattened / Threaded)
Common Sense?
This is something that's built into most modern programming languages.
(reply to this comment) (link to this comment)
Re: Common Sense?
Not the automatic application of the function though...
(reply to this comment) (link to this comment)
Penny Arcade said it best.
Patently Ridiculous
(reply to this comment) (link to this comment)
So. . .
I cant use Trim() anymore? hehe
(reply to this comment) (link to this comment)
If I get myself some trim() -- a la Eddie Murphy in 48 Hours -- can I patent that?
Seriously, the patent filing process is clearly broken in IBM. Really bright new ideas are being passed over in favor of these kinds of "patents" (or at least are being diluted and lost in the noise of all the frivolous filings).
(reply to this comment) (link to this comment)
So does this mean I can't use LTRIM and RTRIM functions is my program without paying IBM a royalty?
(reply to this comment) (link to this comment)
LTRIM, RTRIM, TRIM
are obsoleted by this patent. The patent allows for an automatic L/R/TRIM built into the saving of the data into a database.
However, the patent does not cover displaying the data on a screen in trim'd format. So, if you get user input on one screen, and do not save it to a "database" but display it later on a different screen you do not run afoul of the patent.
I wonder why they didn't include that function too?
Randy
(reply to this comment) (link to this comment)
Re: LTRIM, RTRIM, TRIM
Display is the next patent they will file. You know they get paid for each patant ;)
(reply to this comment) (link to this comment)
Ironic
An add on this page reads, "Want to patent an idea?" www.inventionhome.com.
(reply to this comment) (link to this comment)
Re: Ironic
It's not Ironic...it AdSense. What's ironic is that you read a tech blog and don't understand basic web tech.
(reply to this comment) (link to this comment)
Re: Re: Ironic
It's ironic that there should be an ad for patenting an idea on a web site that repeatedly stresses that ideas aren't and shouldn't be patentable. Or at least I think it is - irony is in the eye of the beholder.
(reply to this comment) (link to this comment)
Prior Art
Should be fairly easy to prove that things like the TRIM() function in BASIC predated this patent. Goofballs!
(reply to this comment) (link to this comment)
Re: Prior Art
The patent isn't for the TRIM function...it is for the automatic use of the TRIM function for data being input into a database.
Not that this is "ground breaking", but the truth of the matter is that no RDBMS that I know of does this today, so they decided to patent it.
They may have done so in order to stop someone else from blocking them (i.e. someone else patenting the idea). It is cheaper to spend resources pushing through a lame patent that you end up owning, rather than having to fight that same patent that someone else "invents".
The only "prior art" that I can think of off-hand is Sybase SQL Server (i.e. MS SQL Server) and its automatic appending of blanks in order to "fill" a column to its fixed size. That is, insert "foo" into a CHAR(10), and the RDBMS automatically appends 7 spaces.
(reply to this comment) (link to this comment)
Re: Re: Prior Art
Of course, if they had just done it and released it, it would BECOME prior art and it would be unpatentable.
(reply to this comment) (link to this comment)
its the numbers game
outright stupid.. except for the numbers game
(reply to this comment) (link to this comment)
So what?
I fail to see how this is important or even a good idea.
If programmers want to trim the whitespace off their strings, they could do so before and they will still be able to do so. What if the programmers want to maintain the whitespace for some reason, will they be able to do so?
(Of course, this is all assuming you're using an IBM database...)
(reply to this comment) (link to this comment)
Does techdirt ltrim()?
(reply to this comment) (link to this comment)
Re:
hhaha they do...prepare to get sued!!!!!
(reply to this comment) (link to this comment)
Re: Re:
Actually, Hypertext markup ignores spaces after the first 2 spaces...
[I type (with "." to represent spaces)
H.....I.....!]
It comes out as:
H I !
(reply to this comment) (link to this comment)
Re: Re: Re:
Does techdirt ltrim()?
I dunno...i don't see the s p a c e s ?
(reply to this comment) (link to this comment)
Two words:
Prior. Art.
Sorry Big Blue!
(reply to this comment) (link to this comment)
I'm not too worried
Big Blue is normally pretty well behaved when it comes to software patents. They certainly aren't trolls. This is one of those defensive patents that prevents someone else from patenting it.
(reply to this comment) (link to this comment)
To Da Lawyers...
Doesn't a professional in the field have to certify that the invention is novel? And if he certifies something that is obviously NOT novel, isn't that perjury?
And if it IS perjury, and the individual IS prosecuted, wouldn't the fear generated cut down on the USPTO's workload?
(reply to this comment) (link to this comment)
Re: To Da Lawyers...
Noone has to certify anything is novel before submitting it as a patent application to the USPTO; that is something they check before issuing the patent (and seem to have failed to do here). An Expert Witness is called to testify only when a patent is being investigated and/or challenged (after its been granted).
(reply to this comment) (link to this comment)
I guess none of you read the actual patent.....
it's about not having to rely on ltrim, rtrim, etc. Any DBA understands that using a function on a table column pretty much sux ass, slows down retreival to a crawl( for any decent sized db). This patent addresses automatically removing them before saving so the dev doesn't have to worry about it on every db update/select. It will be implemented by a DB setting/column setting so it happens automatically without intervention.
I once created an SP that would clean up columns on scheduled basis to trim out uneccesary spaces, to reduce the DB size(yeeears ago when HD space mattered!), took a full weekend to clean up the DB. We still had to be carefule to trim spaces in our code/queries because in between cleanups the spaces would still be there. This method eliminates that.
But to grant a patent.........
(reply to this comment) (link to this comment)
Re: I guess none of you read the actual patent.....
We still had to be carefule to trim spaces in our code/queries because in between cleanups the spaces would still be there. This method eliminates that.
Or... you just trim() before calling your DB update... what is so difficult about that? Hell, I wrote just such a method back when I was getting my Bachelor's degree. it just makes sense.
(reply to this comment) (link to this comment)
Re: I guess none of you read the actual patent.....
Yeah, we got that. Put an Rtrim and Ltrim in a trigger on add/update, and you no longer have to worry about it in the rest of your code. Adding the same functionality directly to the database code and calling it "novel" is nothing less than fraudulent, IMHO.
(reply to this comment) (link to this comment)
Re: I guess none of you read the actual patent.....
The reasoning you use above is why I always thought using the Trim() function before the data was passed to the DB was beneficial in the first place?
(reply to this comment) (link to this comment)
Broken Code
This will break every application that relies on the use of leading and trailing spaces.
(reply to this comment) (link to this comment)
Re: Broken Code
leading and trailing whitespace seems to be an odd thing for code to rely upon.
(reply to this comment) (link to this comment)
Re: Broken Code
Then don't enable the option on those columns. This is an optional property that will be applied to columns by the DBA. If a DBA goes and modifies a schema without testing dependending applications...well, then they aren't a DBA.
(reply to this comment) (link to this comment)
Hey,ifthepatentsystemain'tbroke,don'tfixit!
Where's my money?
(reply to this comment) (link to this comment)
Will the real John Doe please stand up?
Hey, I have been trolling here for a little while now and while I appreciate IBM worrying about the space around me, I don't need them to act as my keeper. There are times I want space between me and the next gal and there are times I don't. But I will decide that, thank you very much!
(reply to this comment) (link to this comment)
Wow. This has to be the worst example of a patent issuing in the face of prior art that I have ever seen.
(reply to this comment) (link to this comment)
Re:
What is the prior art?
Don't say "trim", because that's not what the patent is for.
(reply to this comment) (link to this comment)
Defensive patents
As RevMike pointed out earlier, IBM is pretty good about patenting things just because they can in order to keep someone else from doing it and throwing a spanner in the works. They then allow free license to use the patent. Until software patents are finally and properly disallowed, this kind of behavior is to be expected. It keeps the trolls from making a mess of things.
The frightening part is if IBM's management changes and their attitude about such things changes with it. They could be holding a bajillion patents that would make writing software nearly impossible. For instance, MIT currently holds a patent on linked lists that was just awarded a couple years ago. Those have been in use since the 1960s, and tons of prior art exists because it's such a fundamental methodology. They're not enforcing their patent, but what if they did? Tons of software across almost all disciplines would suddenly be infringing.
(reply to this comment) (link to this comment)
Obvious and trivial can't be patented
This is SOOOO obvious, and there is certainly prior art for this. Oracle's varchar data types automatically right-trim input data. Having it left-trim input data as well could be done as a trigger or trivial update to the database code. In my opinion, this is patent proof that the US Patent Office's review process for software patents is totally fubar (pun intended)!
(reply to this comment) (link to this comment)
Dumb
So does this mean that I start to infringe on IBM's patent with code like this:
char *p = string;
while ( isspace(*p++))
;
Building a trivial function into something doesn't seem patent material to me.
I thought patents were granted for "non-obvious" inventions.
(reply to this comment) (link to this comment)
Add Your Comment