While I fully agree that we should "[r]espect the work required to produce new ideas," etc., and giving proper credit for work done by others seems to me basic academic honesty, I feel that in the details this item starts out from wrong principles.
As a note before I get into this, some of this may sound at first glance like extreme libertarianism or something like that. If it comes across that way, please try re-reading it again because I do not agree at all with extreme libertarianism and I think that IP laws can serve a socially useful purpose, though of course they can also do harm as well as good.
What we now call "intellectual property laws" are government-enforced restrictions on freedoms. They're somewhat unusual because, unlike restrictions against theft and the like, where you deprive someone of the use of something, they disallow the use of something that would otherwise cost nobody else anything at all. (In economic terms, they're restrictions on the use and taking of non-rvial goods.) For example, if I print copies of a book you wrote and distribute them, you do not lose anything you already had; you can still do whatever you like with your book as well and you're not forced to pay for my printing or distribution of the book.
Societally, this is the ideal state for existing works. Progress is made by building on the work of others, and clear harm can be done to or progress as a society without this. Think of a case, for example, where someone spent several years working out a difficult mathematical proof that would be useful for my work; we can easily see the societal harm that results if I'm not allowed to use that proof in my work because it's "owned" by someone else.
Originally, patents were exactly this: a grant from the king of a restriction on doing something that allowed certain people to profit from doing that thing because others weren't allowed to do it. Economists call this "rent seeking" and it's generally considered to be a bad thing; these days most people and societies agree that we shouldn't do this.
However, later another completely separate idea arose: that it would improve our society to have more intellectual property created, and we could encourage this by somehow rewarding the people who created it. One mechanism proposed for the reward is that for a period of time we restrict the normal rights of others to use certain ideas, expressions or whatever, allowing the creators to seek rent, but this harm would be more than counterbalanced by, in the long run, having more free works floating around in our society for anyone to build on. It's very important to note there that at its core this is not about giving or asserting creator's rights to their non-rival creations; that's just a mechanism to support the true aim.
And that's why I find a statement like this objectionable:
I'd say, no, this is not right. What we're really recognizing here is that we should limit the amount we block the (re-)use of non-rival goods. Our default starting point for thinking about this should not be "creators have rights" but "to what degree do we feel comfortable taking away the natural right to the use of any ideas you come across."
The examples you want to be looking at are things like, "what if QuickSort were patented right now?" (Note that it does not have to be correctly patented; you're free to assume it's the case that the patent should not have been allowed, and would fall to a challenge if someone were willing to spend enough money and time in court.) While this example is simplified for clarity, we deal with this sort of thing all the time in the computing field, from Amazon's "one click" patent to patents on ways of compressing data.
As a side note, also be careful with how you think about "open source." A great deal of open source software is written by people who are committed to the idea of greatly restricting what use people may make of that software. (The GPL license is the most common way of doing this.) This isn't always clear due to the (backwards) claims about "free as in speech vs. free as in beer" (the GPL is aimed towards the latter), but it does come out in other things that say, such as Bruno Haible's reasoning for using the GPL for libgmp, which means if you want to use that library in your software you must give away all your source code for free: "Building libgmp.a was just too hard work. Other people shouldn't get it for free."