Question: Is my credit score good enough for a car loan?

Hi there. I am trying to take a car loan out for a car, and was curious if anyone thinks I would get approved for one-before I take the time to go and apply and add inquiries to my report.
My current credit score is between 676-690 between the three credit agencies. I currently have about 800 in balances. But have never had anything in collections. My issue is that I I only have 1 year and 3 months worth of credit history. What is the likelihood that I will get approved? my income is about 3300 a month.

Will grammar errors become correct after enough people use them for enough time?

First let me state the obvious—based on my own experience—that hordes of people are confused about certain basic grammar principles. For example, I so often see mistakes in choosing the pronoun to use with direct objects, especially when introduced by and with another person, as in “Joe talked about the book with Anna and [I/me].” Another one is the near 100% usage of “lay” instead of “lie” (and “laid” instead of “lay”) nowadays, even in published sci-fi books that most certainly had professional editors proofread them.

Do you see these kinds of changes as predictors of the future? Will they become correct after enough people have used them long enough? Both of these problems drive me bonkers, and I fight my own little battle against them when possible.

I know that dictionaries are descriptive rather than prescriptive. They’re technically history books. But grammar and style books seem less so, to me. Those seem as though they ought to be prescriptive in order to reduce the erosion of structured and meaningful language.

In this tension between “what people are doing” and “what people should be doing,” are we doomed to forever bear these errors just because too many people don’t know the proper way? Is it worth fighting against them? Do I capitulate to the people who are quick to accept any mass public usage as a fait accompli?

Update

I don’t imagine for a second that my own writing is error free. The difference with me is that when I do learn I have something wrong in my mind, I immediately change once I clearly understand it.

I also am not imagining that language is static or ignorant that today’s correctness is all too often yesterday’s error. The whole point of my question here, then, is: is there value in slowing the change, and if so, how is that done and how effectively?

Another way to look at what I’m trying to ask is that in a way I am hoping to define the limits of pedantry. If being pedantic is slavish adherence to outdated rules in the face of actual and foregone changed reality, then when do we conclude that a change is a foregone conclusion?

Poking fun at my own errors to highlight supposed pedantry on my part is to miss everything I am trying to say.

Update 2

It may not be an easy question, but since language change actually does or does not happen, and every person acts with more or less intention in regards to language change, there must be an answer. What I wanted to explore is the value in efforts to teach the “proper way” vs. the opposite end of the spectrum where any new usage is not just accepted but welcome or even sought out. You know, I’m reminded of something: the conservative/liberal scale:

   |------+------+------+------+------+------|
radical              moderate             radical
liberal                                 conservative

The funny thing about those at the ends of the spectrum is that they both want things to change. The radical liberal wants language to change to something new quickly, just for the sake of newness or evolution or some other not-necessarily-realistic ideal. Similarly, the radical conservative wants things to change just as much, but back to the way it used to be, just for the sake of sameness or continuity or some other just-as-not-necessarily-realistic ideal.

I’m not particularly asking where we should be on this scale as that’s purely subjective. I’m trying to ask about something purely practical and real-world: what is actually effective? Is language change inevitable, and how fast? Are efforts to reduce its rapid morphing either effective or worthwhile in any measure? All these things considered, what position with respect to language evolution is livable, practical, and sensible?

Feel free to edit my grammar. No comment necessary.

Update 3

Something that may be of interest to both my supporters and detractors: this ‘Kinetic Typography’ video by Stephen Fry. I enjoyed it. I don’t disagree with him, but I am not sure I am wholehearted in this lack of disagreement.

Question: Have you had enough of these Democrat ‘immigrant’ caravans? Are you really to VOTE OUT ALL…

joedlh — Dems don’t have to compel them to march north, just ENABLE them anonymous — The Democrats have ORGANIZED the caravan. You need to spend more time working on your reading comprehension instead of trying to twist everything into a slam against Republicans. tham153 — Your STUPIDITY is showing.

Transforming proportion data: when arcsin square root is not enough

Is there a (stronger?) alternative to the arcsin square root transformation for percentage/proportion data? In the data set I’m working on at the moment, marked
heteroscedasticity remains after I apply this transformation, i.e. the plot of residuals vs. fitted values is still very much rhomboid. Thanks in advance for any suggestions.

Edited to respond to comments: the data are investment decisions by experimental participants who may invest 0-100% of an endowment in multiples of 10%. I have also looked at these data using ordinal logistic regression, but would like to see what a valid glm would produce. Plus I could see the answer being useful for future work, as arcsin square root seems to be used as a one-size-fits all solution in my field and I hadn’t come across any alternatives being employed.

Bootcamp: The startup disk does not have enough space to be partitioned

When clicking continue on the first Boot Camp screen, I am presented with a dialog that displays:

The startup disk does not have enough space to be partitioned.
You must have at least 39 GB of free space available.

Screenshot of Bootcamp

This is despite “About this Mac” reporting 271GB of free space:

Screenshot of About this Mac

And “Disk Utility” reporting 268GB of free space:

Screenshot of Disk Utility

Searching online I found this reddit thread which suggested it was the result of Time Machine which i recently enabled and suggests to run sudo tmutil thinlocalsnapshots / 999999999999 which I did, it ran sucessfully, and I restarted, yet to no avail.

What is the solution to this instance of Apple’s poor engineering?

Running macOS High Sierra 10.13.4 (17E202)

Could a species with a lifespan of 20 years achieve enough technological progress to reach space?

I’m current planning a story which includes a race of diminutive humanoid aliens, one of whom crash-lands on Earth. They’re fully sentient, and have the same average intelligence level as humans. The other main difference from humans, besides their height, is that their lifespans are much shorter.

Using sols (Earth years) as a comparison: an average human reaches maturity around their mid-to-late teens, and has a life expectancy of 70-80 sols. An average member of the alien species reaches maturity at around 4 sols, and has a life expectancy of about 20. 30 is almost unheard of. The crash-landing alien is therefore astonished that the first human he runs into is 36 sols old.

This raises an interesting problem. 20 years isn’t an awfully long time to do anything at all. If they spend the same number of sols in education as we do, they won’t have much time to do anything with that education by the time they graduate. If they spend the same proportion of their lives in education as we do, that’s about 3-4 sols, which is nowhere near enough time to actually learn a whole lot.

So I imagine, compared to humanity, these relative mayflies are going to find technological progress quite difficult. It won’t be so much “standing on the shoulders of giants” as “balancing precariously at the top of a chain of fifty aliens, all standing on each others’ shoulders”. So I ask you: could a species whose average livespan is only 20 years ever progress far enough to reach space?

  • By “reach space”, I mean a program comparable to the Apollo missions – if they can get that far, I’ll make the bold assumption that they will eventually reach interstellar travel.
  • You may assume that their resources, planetary environment, and desire are all sufficient for space travel.
  • You may assume that their discoveries are sufficiently documented to be remembered and built upon, as mankind’s discoveries (generally) have been.
  • You may assume that the aliens do not have the means to artificially extend their lifespans (be it via DNA modification, cybernetic enhancement, or whatever).

I’m aware of this question, but there, the “shorter lifespan” is that of humanity. I’m going 3-4 times shorter than that, and I want to know how ”far” they could progress, not how fast.

ValueError:not enough values to unpack (expected 2, got 0)

based on python3.6 and
my dataset graph.txt is like below:
enter image description here

and each line split by t

my code:

   text_file, graph_file = self.load(text_path, graph_path)

   self.edges = self.load_edges(graph_file)

def load(self, text_path, graph_path):
    text_file = open(text_path, 'rb').readlines()
    graph_file = open(graph_path, 'rb').readlines()
    return text_file, graph_file

def load_edges(self, graph_file):
    edges = []
    for i in graph_file:
        edges.append(map(int, i.strip().decode("utf-8").split('t')))
    return edges

however when i run below function :

def negative_sample(self, edges):
    node1, node2 = zip(*edges)
    sample_edges = []
    func = lambda:self.negative_table[random.randint(0,config.neg_table_size - 1)]
    for i in range(len(edges)):
        neg_node = func()
        while node1[i] == neg_node or node2[i] == neg_node:neg_node = func()
        sample_edges.append([node1[i], node2[i], neg_node])
    return sample_edges

I getting a value error:In function:negative_sample node1, node2 = zip(*edges)
ValueError: not enough values to unpack (expected 2, got 0)