Well, in the district where my mom is a school psychologist, she is essentially responsible for about 4-5000 students spread across 4 schools. Part of her job is to screen kids for special education. Most of her day is spent convincing teachers in various schools that Timmy doesn't have a learning disability just because he acts out occasionally, and that even if he did a Special Ed classroom may well cripple him for the rest of his life. She only has about a day per week per school to administer assessments, provide classroom seminars, and manage confused teachers.
Librarians have a similar problem. Many districts in Washington can't afford even one per school anymore. The library in those districts is either closed or at reduced capacity if the librarian is not in.
I voted for I1135 mainly because it would increase funding for classified staff again, and not because it reduced class sizes.
Then this article is a poorly written piece. It makes no mention of that. I'm not in WA so I've no clue what was in that initiative. But the article sure does make it sound like it was just for reduced class sizes. I feel like the author is against it and her bias is seeping through.
The problem I see with using EITC in this way is that it masks the cost of labor for any company paying minimum wage. A huge hole in arguments against raising the minimum wage is that few seem to address the fact that many workers are already receiving additional pay in the form of benefits. The benefits are more expensive and less flexible than an equivalent pay raise, because new bureaucratic infrastructure inevitably has to be created to manage the new benefit. Is that really more efficient than just paying higher wages?
Practically every bank I do business with has these limitations. By contrast, my gmail password is 24 characters long, and contains capital and lowercase letters and punctuation. It makes me wonder what their back-end systems look like that they can't handle passwords longer than that.
My understanding was that although regular CGI respawns the command interpreter every time the web server is queried, FastCGI reuses the same command interpreter and just updates the environment.
Most of the time, change is effected through long effort and reflection, and by the end you don't see you've changed at all.